Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Abstracts from the International Society for Therapeutic Ultrasound Conference 2017

Abstracts from the International Society for Therapeutic Ultrasound Conference 2017 Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 https://doi.org/10.1186/s40349-018-0110-x MEETING ABSTRACTS Open Access Abstracts from the International Society for Therapeutic Ultrasound Conference 2017 Nanjing, China. 31 May - 02 June 2017 Published: 21 May 2018 RESULTS 18F-DPA-714 uptake was increased at the sonication Oral Presentations sites in all locations (Fig. 1). The ratio of the percent increase in SUV between pFUS+MB treated striatum and hippocampus to O1 contralateral side is depicted in graph (mean+/-SEM) clearly Neuroinflammation after disrupting the blood brain barrier with showing large increase in uptake for both regions compared to pulsed focused ultrasound and microbubbles imaged by 18F-DPA- normal brain. The neuroinflammatory changes persisted for at 714 PET and MRI least 14 days after 2 weekly sonications. The coefficient of vari- Zsofia I. Kovacs, Georgios Z. Papadakis, Tsang-Wei Tu, Sanhita Sinharay, ation for PET scans was <10%. This corresponded to Iba1 activa- William C. Reid, Bobbi Lewis, Dima A. Hammoud, Joseph A. Frank tion visible on histology. Figure 2 contains normalized National Institutes of Health, Bethesda, Maryland, United States histograms from VOI for Group 2 and Group 3 rats derived from Correspondence: Zsofia I. Kovacs pFUS+MB treated (ipsilateral) and contralateral brain that shows Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O1 a shift to lower T2* values for sonicated regions. OBJECTIVES Blood brain barrier (BBB) disruption with MR-guided pulsed focused ultrasound (pFUS) and microbubbles (MB) has CONCLUSIONS Rats receiving pFUS+MB to open the BBB showed a been advocated as a noninvasive adjuvant treatment for malig- clear upregulation of TSPO expression consistent with microglial nancies and neurodegenerative diseases. A sterile inflammatory activation/neuroinflammation, even after one sonication session. reaction has been recently described in the brain as a result of Histograms derived from T2* maps MRI clearly shows that sonication pFUS+MB (Kovacs et al. 2016). However, one potential issue of with BBB algorithm results in left shift in T2* values that would be weekly pFUS+MB treatments is the lack of data on the long- consistent with hypointense voxels on T2*w MRI and abnormatilies term effects on inflammation. The purpose of this study was to on histolology. These preliminary results contradict current assump- evaluate the effects of multiple weekly courses of pFUS+MB tions that the effects of pFUS+MB are confined primarily to the endo- exposures in the rat brain using micro-Positron Emmision Tom- thelium and vessel wall. Further assessment of the long-term effects ography (PET) and 18F-DPA-714, a marker of translocator pro- of pFUS+MB is necessary before this approach can be widely imple- tein (TSPO) upregulation/microglial activation and an indication mented in clinical trials. of neuroinflammation. METHODS Female rats were assigned to three different groups References based on the number of weekly pFUS+MB: Group 1: pFUS+MB Kovacs, Z. I., et al. (2017). "Disrupting the blood-brain barrier by focused ultrasound x1, PET scans performed 24 hours later (n=6) ; Group 2: pFUS induces sterile inflammation." Proc Natl Acad Sci U S A 114(1): E75-E84. +MB x2 with PET scans performed within 10 days after 2nd son- O'Reilly, M. A. and K. Hynynen (2012). "Blood-brain barrier: real-time feedback- ication (n=5) ; and Group 3: pFUS+MB x6 with PET scans per- controlled focused ultrasound disruption by using an acoustic emissions- formed 7-9 days later (n=5). The left striatum (str) and right based controller." Radiology 263(1): 96-106. hippocampus (hc) were targeted in all animals. 100 μlof MB (OptisonTM, GE Healthcare, Little Chalfont, UK) was administered intravenously over 1 minute starting 30 secs before pFUS. Acous- tic energy was delivered to the brain using “BBB configuration function” based on algorithm reported (O’Reilly et al. 2012) to de- and termine optimal acoustic pressure for BBB opening via 1.5f 2.5f ultra harmonic acoustic emission detection for every single pulse (9 focal points, 120 sec/9 focal points – striatum, 120 sec/4 focal points – hippocampus) using an 825 kHz hydrophone with a single-element spherical FUS transducer (center frequency: 589.636 kHz; focal number: 0.8; aperture: 7.5 cm; RK-100, FUS Instruments, To- ronto, Ontario, Canada). T2* map were created from multiecho gradient echo sequence at 3T (Achieva, Philips Healthcare, Andover, MA) through the rat brain with TE=7 msec, echo train length 5 and echo spacing 7 and Tr=1500 msec. T2* maps were created by fitting signal intensity at each voxel to a single exponential fit with in-house software Fig. 1 (abstract O1). 18FDPA714 PET scans (coregistered to MRI and histogram analysis was performed on volume of interests (VOI). template) in two rats: A: Hippocampus uptake 24 hours after one Static microPET/CT scans emission data was acquired 30-60 min after sonication. B: Striatum uptake 10 days after six weekly sonications. C: injection of 18F-DPA-714. VOIs were drawn in the targeted areas and Graph of % difference between sonicated ipsilateral and contralateral uptake was compared to the contralateral unaffected side. Uptake striatum and hippocampus regions for Groups 1, 2 and 3 values were normalized to cerebellum. © The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 2 of 122 fiber structure- and gray matter-abnormalities on DTI MRI were de- tected in regions with the T2* abnormalities suggestive of increased astrogliosis and transient axonal damage. MRI findings following pFUS+MB x6 demonstrated more pronounced evidence of damage within the parenchyma and atrophy. FDG-PET did not show differ- ences between sonicated and contralateral cortex or hippocampus at any time point. BrdU showed evidence of increased neurogenesis in pFUS+MB treated regions. CONCLUSIONS The long term effects of pFUS+MB exposures in rats revealed that both single and repeated pFUS+MB cause structural in- jury at the location of sonication up to 13 weeks post treatment based on advanced imaging techniques. Histological evidence Fig. 2 (abstract O1). Mean normalized histograms derived from VOI showed that associated with the pathological changes observed by from pFUS+MB treated cortex/striatum and hippocampus (ipsilateral) MRI, there was evidence of neuronal damage and loss, neurogenesis compared to contralateral brain forGroup 2 and 3 cohorts of rats. and activated microglia. These results suggest the importance of There is clear shift to lower T2* values for sonicated regions for 2x v long term monitor of the brain following low intensity pFUS+MB be- fore its clinical translation. O2 Long term effects of pulsed focused ultrasound and microbubbles References detected by multivariate imaging modalities Arvanitis, C. D., et al. (2016). "Cavitation-enhanced nonthermal ablation in Zsofia I. Kovacs, Tsang-Wei Tu, Georgios Z. Papadakis, William C. Reid, deep brain targets: feasibility in a large animal model." J Neurosurg Dima A. Hammoud, Joseph A. Frank 124(5): 1450-1459. National Institutes of Health, Bethesda, Maryland, United States Downs, M. E., et al. (2015). "Long-Term Safety of Repeated Blood-Brain Barrier Correspondence: Zsofia I. Kovacs Opening via Focused Ultrasound with Microbubbles in Non-Human Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O2 Primates Performing a Cognitive Task." PLoS One 10(5): e0125911. OBJECTIVES Blood brain barrier (BBB) opening by Guided Pulsed Fo- cused Ultrasound (pFUS) and microbubbles (MB) is a non-invasive O3 treatment of various central nervous system diseases. However, the Characterization of different microbubbles in assisting focused potential adverse effects of repeated pFUS+MB exposure have not ultrasound-induced blood-brain barrier opening 1 2, 3 3, 4 5 been thoroughly elucidated and may limit clinical translation. To date Sheng-Kai Wu , Po-Chun Chu , Wen Yen Chai , Shih-Tsung Kang , 3 5 5 3 MRI scans of repeated BBB opening by pFUS+MB have been Chih-Hung Tsai , Ching-Hsiang Fan , Chih-Kuang Yeh , Hao-Li Liu achieved without hemorrhage, edema and behavioral changes in Institute of Biomedical Engineering, National Taiwan University, Taipei, non-human primates (Arvanitis, et al. 2016; Downs, et al. 2015). By in- Taiwan; Department of Research and Development, NaviFUS corp, corporating detailed multimodal imaging, we characterized the long Taipei, Taiwan; Department of Electrical Engineering, Chang-Gung term effects of single or repeated pFUS+MB in the rat brain. The pur- University, Taoyuan City, Taiwan; Department of Diagnostic Radiology pose of the study is to reveal the morphological and pathological and Intervention, Chang-Gung Memorial Hospital, Taoyuan City, Taiwan; changes following repeated BBB opening in the striatum and hippo- Department of Biomedical Engineering and Environmental Sciences, campus as monitored by 3T and 9.4T MRI, FDG-positron emission National Tsing Hua University, Hsinchu, Taiwan tomography (PET) and histology over 13 weeks. Correspondence: Sheng-Kai Wu TM METHODS pFUS+MB (Optison , GE Healthcare, Little Chalfont, UK) Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O3 0.3 – 0.5 MPa, 10 ms burst length, 1 % duty cycle, 9 focal points, 120 sec/9 focal points – striatum, 120 sec/4 focal points – hippocampus, This abstract is not included as it has already been published: 589.636 kHz; focal number: 0.8; aperture: 7.5 cm; (FUS Instruments, Wu S-K, Chu P-C, Chai WY, Kang S-T, Tsai C-H, Fan C-H, Yeh C-K, Liu H-L. Toronto, Ontario, Canada) was targeted in female rats (n=6/group) Characterization of Different Microbubbles in Assisting Focused either once or six weekly to the striatum and the contralateral Ultrasound-Induced Blood-Brain Barrier Opening. Sci Rep. 2017; 7. Available hippocampus. 100 μL of MB were administered intravenously over 1 from: http://www.nature.com/articles/srep46689 doi:10.1038/srep46689 minute starting 30 secs before pFUS. Rats received 3 daily doses of 300mg/kg 5-Bromo-2′-deoxy-uridine (BrdU, Sigma-Aldrich, St. Louis, O4 MO) intraperitoneally before sonication to label proliferating cells Development of an A-Synuclein (SNCA)-based mouse model for in vivo. T2, T2* and Gd-enhanced T1-weighted images were obtained Parkinson's disease by ultrasound-guided CNS delivery by 3.0T MRI (Achieva, Philips Healthcare, Andover, MA), T2, T2* and 1 2 2 Chung-Yin Lin , Yu-Chien Lin , Hao-Li Liu diffusion tensor imaging (DTI) were performed by 9.4T MRI (Bruker, Institute for Radiological Research, Chang Gung University, Taoyuan Billerica, MA). Parameters for DTI: 3D spin echo EPI; TR/TE 700 ms/37 2 2 City, Taiwan; Department of Electrical Engineering, Chang Gung msec; b-value 800 sec/mm with 17 encoding directions; voxel size University, Taoyuan City, Taiwan 200 μm (isotropic). Fractional anisotropy (FA) and the asymmetry of Correspondence: Chung-Yin Lin magnetization transfer ratio (MTRasym) were derived for mapping Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O4 structural injury and glucose levels. Rats received ~1.1 mCi of 18F- FDG via intravenously to quantitate glucose uptake by PET/CT OBJECTIVES Parkinson’s disease (PD) is the second most common (Inveon, Siemens, Munich, Germany). PET emission data was acquired neurodegenerative disease characterized by the progressive degen- for 60 min. Dynamic images were reconstructed and image analyses eration of dopaminergic neurons in the substantia nigra (SN) and the was performed (PMOD Technologies Ltd., Zurich, Switzerland). presence of α-synuclein-containing inclusion bodies in the cytoplasm Animals were euthanized 7 or 13 weeks after the first pFUS of neurons. In this study, we propose to develop a novel asynuclein treatment. Histological evaluation of brain and tracking of BrdU over-expression PD mouse model via ultrasound-guided CNS delivery tagged cells was performed. of SNCA gene. RESULTS Gd-enhanced T1-weighted images after each sonication METHODS A plasmid encoding both the green fluorescent protein demonstrated BBB disruption in the striatum and the hippocampus (GFP) gene and the SNCA gene was prepared. A gene-liposome sys- at 3T. Gd T1 enhancement, T2 and T2* abnormalities were not seen tem, in which the liposomes are designed to carry SNCA plasmid in the brain 1-day post pFUS+MB at 9.4T MRI. Hypointense regions DNA, forms liposomal-plasmid DNA (LpDNA) complex. Ultrasound appeared on T2* MRI 2 weeks post pFUS+MB consistent with devel- (US) exposure used the SonoPore KTAC-4000 to induce BBB opening opment of microhemorrhages within the parenchyma. White matter Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 3 of 122 (1.0-MHz, voltage = 85 V, burst length = 10 ms, 10% duty cycle, and optimization procedure, from the experimentally determined Dox 3-min sonication duration). The longitudinal expression of GFP was kinetics under the different conditions tested. Finally, based on quantitated via an in vivo imaging system (IVIS). The SNCA gene ex- the fitted values, a sensitivity analysis was performed and the pression level was confirmed via immunoblotting, and histological most important parameters that affect the Dox transport and staining will be used to identify transfected cells via fluorescent mi- cellular uptake in the tumor interstitium were determined. croscopy. Dopamine and metabolic DOPAC protein levels in the RESULTS Multiphoton microscopy revealed up to one order of mag- brain were determined via HPLC. nitude higher Dox extravasation after FUS- BBB/BTB disruption as RESULTS With longitudinal observation of IVIS monitoring, animals compared to control. In addition, a five-fold increase of Dox penetra- with US treatment showed significant promotion of LpDNA release tion was found after FUS- BBB/BTB disruption compared to control into the SN and demonstrated enhanced expression of genes upon (>100μm vs. <20μm, based on Dox penetration regression). The nu- sonication with US-BBB opening, while both the GFP and α-synuclein merical model indicated that only the vessel diffusion coefficient 2 2 protein expression were successfully measured via Western blotting. (4.71 ± 1.7μm /s Vs 0.41 ± 0.1μm /s, a tenfold increase) and the Immunoblotting and histological staining confirmed the expression pressure difference (4.3 ± 0.45mmHg vs 2.9 ± 0.32mmHg) were of reporter genes in neuronal cells. significantly different between the FUS- BBB/BTB disruption and the CONCLUSIONS This study suggests that IV administration of LpDNA control. While the increase in vessel diffusion coefficient was in combination with US-BBB opening can provide effective gene de- anticipated, the increase in pressure difference, which lead to a livery and expression in the SN, demonstrating the potential to system’s Peclet number greater than one (Pe>1), suggested a achieve non-invasive and targeted gene delivery for α-synuclein fundamental change in the interstitial transport mechanism. over-expression PD mouse model. Assessment of the temporal evolution of the drug concentration in the interstitial space in the experimental data verified that the transport after FUS-BBB/BTB disruption was convection dominated. O5 Single cell analysis revealed significant Dox uptake by endothelial cells Experimental and numerical assesment of Doxorubicin (EC), suggesting that microbubble vibrations can lead to significant pharmacokinetics in brain metastasis from breast cancer after changes in cellular transmembrane transport. Interestingly, in the FUS focused ultrasound-induced blood-brain/blood-tumor barrier treated animals’ stroma cells (SC) appeared to take-up the drug at disruption higher rate than the endothelial cells (uptake slope 0.44 vs 0.93 in EC 1,4 2 1 2 Costas Arvanitis , Vasileios Askoxylakis , Yutong Guo , Jonas Kloepper , and SC respectively). These differences in the uptake curves led to 2 3 2 5 Meenal Datta , Miguel Bernabeu , Dai Fukumura , Nathan McDannold , significant changes in the rate of cellular transmembrane transport in Rakesh Jain the numerical model during parameter fit. Finally, sensitivity analysis Mechanical Engineering, Georgia Institute of Technology, Atlanta, showed that FUS-BBB/BTB disruption makes drug uptake more Georgia, USA; Radiation Oncology, Massachusetts General Hospital, sensitive to all the parameters of the system, suggesting that the main Harvard Medical School, Boston, Massachusetts, USA; Centre for Medical barrier to drug transport has been overcome. Interestingly, the BBB/BTB Informatics, University of Edinburgh, Edinburgh, UK; Biomedical permeability remains an important parameter of the system even after Engineering, Georgia Institute of Technology, Atlanta, Georgia, USA; FUS, indicating that further disruption may be beneficial. Radiology, Brigham and Women's Hospital, Harvard Medical School, CONCLUSIONS The in vivo and in silico data demonstrate signifi- Boston, Massachusetts, USA cant changes in the tumor microenvironment after FUS-BBB/BTB Correspondence: Costas Arvanitis disruption. The most notable changes included: i) increase in Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O5 BBB/BTB permeability, ii) transition to convection dominated drug transport, and iii) increased cellular transmembrane transport in OBJECTIVES Blood brain and blood tumor barriers (BBB and BTB) endothelial cells and stroma cells. Sensitivity analysis showed that constitute a major obstacle to the transport of therapeutics in brain the system has become more amenable to interventions, suggest- tumors. Focused ultrasound (FUS), when combined with circulating ing that FUS can lead to the development of new therapeutic microbubbles, provides a noninvasive method to locally and transi- strategies to treat brain tumors. ently disrupt the BBB/BTB. While several studies have demonstrated its potential for targeted drug delivery, there is a lack of fundamental understanding of the impact of this method on the pharmacokinetics O6 of anticancer agents in the brain microenvironment. In this study, we Acoustic emissions during blood-brain barrier disruption with examine the impact of FUS-induced BBB/BTB disruption on the trans- focused ultrasound and real-time feedback control under infusion port of the chemotherapeutic agent doxorubicin (Dox) in a model of administration of microbubbles – feasibility study in rodent model 1 1 1 1 breast cancer brain metastases using intravital microscopy and drug Chenchen Bing , Debra Szczepanski , Imalka Munaweera , Yu Hong , 2 1,2 transport mathematical modeling. Ian Corbin , Rajiv Chopra METHODS Human HER2-amplified and estrogen dependent BT474 Radiology, UT Southwestern Medical Center, Dallas, Texas, USA; breast cancer cells, genetically modified to express green fluorescent pro- Advanced Imaging Research Center, UT Southwestern Medical Center, tein, were stereotactically implanted in the brain of mice with cranial win- Dallas, Texas, USA dows.Atatumorsizeof~20-40mm , BBB/BTB disruption was performed Correspondence: Chenchen Bing using FUS exposures by a custom built portable FUS system and Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O6 concurrent i.v. administration of microbubbles. To cover the entire tumor and its periphery four non-overlapping sonications were OBJECTIVES Glioblastoma multiforme (GBM) is the most lethal pri- performed. Shortly after sonication, the auto-fluorescent mary brain tumor, with aggressive and fatal progression of disease chemotherapeutic agent Dox (7.5mg/kg) was administered i.v. inevitable. During the earlyinfiltration of GBM cells into normal brain The pharmacokinetics of Dox, including intratumoral uptake and regions surrounding the primary tumor, this peritumoral environ- clearance, were measured for 15 minutes using intravital ment has an intact blood-brain barrier (BBB) which inhibitsthe deliv- multiphoton microscopy. The Dox cellular kinetics were also ery of therapeutic agents. BBB disruption with focused ultrasound determined. For mathematical analysis, we developed a numerical could be used to achieve peritumoral delivery of therapeutic agents model combining the Navier-Stokes and Brinkman equations for flow to tackle infiltrative cancerprogression. However, variability in vascu- modeling in the tumor vasculature and interstitial space, coupled with lature structure and function around a tumor leads to heterogeneous a convection-diffusion-reaction model of drug transport based on perfusion of microbubbles which can impact the acousticthresholds Michaelis-Menten kinetics. The model parameters (vessel permeability, required for BBB disruption. A more reliable means of BBB disruption interstitium porosity, rate of cellular transmembrane transport, tissue is desired. Prior studies suggest that by integrating an acoustic emis- hydraulic conductivity, and pressure difference driving flow across the sion detector and afeedback control algorithm to evaluate the inten- vessel wall and interstitium) were inferred, using a numerical sity of acoustic emission at sub-/ultra-harmonics, more controlled Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 4 of 122 and consistent BBB opening can be achieved. The goalof this study O7 was to evaluate the acoustic emissions and BBB opening in rodents Unilateral focused ultrasound-induced blood-brain barrier opening and validate the feasibility of real-time feedback control under intra- redistributes hyperphosphorylated Tau in an Alzheimer's mouse venous infusionadministration of Optison microbubbles. model 1 1 1 3 METHODS A custom-built focused ultrasound transducer with a central Maria Eleni Karakatsani , Tara Kugelman , Shutao Wang , Karen Duff , 1,2 frequency of f = 0.5MHz was attached to a stereotactic system and Elisa E. Konofagou used to deliver ultrasoundenergy into the target brain region. One Biomedical Engineering, Columbia University, New York, New York, USA; 2 3 piezocomposite hydrophone with resonant frequency at 0.75MHz was Radiology, Columbia University, New York, New York, USA; Pathology built to acquire the signals emitted from stimulatedmicrobubbles. A and Cell Biology, Columbia University, New York, New York, USA feedback control algorithm was implemented in LabVIEW to quantify Correspondence: Maria Eleni Karakatsani the area under curve (AUC) within sub-/ultra-harmonic bands during Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O7 theultrasound exposure and to adjust the focal pressure accordingly based on the difference between current AUC and a desired threshold. OBJECTIVES Focused ultrasound has been shown to interact with Initial in vitro tests wereperformed in which microbubble was infused Alzheimer’s pathology and particularly to trigger a mechanism that into a single tube (0.08 mL/min, tubing I.D. = 1mm), and the AUC was results in the reduction of the amyloid plaque load. However, a less quantified at 0.5f ,1.5f and 2.5f as a function offocal pressure and studied interaction is that of ultrasound with the tangle formation 0 0 0 microbubble concentration. Due to the significant response detected at that has been implicated in the cognitive decline of Alzheimer’s pa- 1.5f (0.75MHz), this specific frequency band was used for controlling in tients. Tau pathology can be characterized by increased density of thefollowing experiment. The ability to maintain acoustic emissions at a the hyperphosphorylated tau protein that results in tangle formation. target AUC level was also evaluate in vitro. Next an in-vivo study was At the early stages of Alzheimer’s disease, tau protein can be local- performed in a rat model toevaluate the acoustic emissions and the ized primarily in the axons while in late pathology, somatodendritic feasibility of real-time control of the AUC at a target level. Acoustic tau is more pronounced. With the current study we investigate the emissions from a bolus injection and continuous infusionwere evalu- interaction of focused ultrasound-induced blood-brain barrier open- ated. For bolus injection, a fixed pressure level of 0.54 MPa was applied, ing with the tau distribution. Moreover, the unilateral sonication of while for infusion experiment, the feedback control was used to control the transgenic brain provides a unique opportunity to explore poten- the AUC atvarious levels. Evans blue dye was used as an indicator of tial bilateral effects. BBB opening. METHODS For this study the initial cohort included 10 mice of the RESULTS A minimum transmit focal pressure of 0.33 - 0.41 MPa was rTg4510 line (3.5 months old), 5 of which were randomly assigned to required to trigger the bubble activity in the sub/ultra-harmonic the control group that did not receive any sonication and 5 to the bands in vitro, with the greatestchanges occurring at 0.75 MHz (1.5f ) treatment group. The treatment group received a double sonication (Fig. 1B). With varying concentration of microbubbles, the feedback covering almost the entire hippocampal region once per week for 4 control algorithm could adjust the focal pressure accordinglyto consecutive weeks. The day after the last sonication the mice were achieve a consistent harmonic emission (Fig. 1C). By observing the sacrificed. The brains were sectioned and counterstained for tau pro- in-vivo AUC response of bolus injection used previously to success- tein (AT8) as well as microglia activation (CD68). The images were ac- fully open the BBB, a thresholdAUC value was selected at 1.5f and quired by means of confocal microscopy over a z-series to account tested with a continuous infusion of microbubbles. Successful main- for depth differences and enabling co-visualization of the tau protein tenance of the AUC at different target value was achieved invivo at and microglia distribution. A custom algorithm was constructed to multiple locations in the brain, and BBB opening was confirmed as quantify the number of cells and the axonal distribution of the tau- leakage of Evans Blue at the target locations (Fig. 1D). marker. Background noise was automatically removed by color-based segmentation using k-means clustering and the cells were detected CONCLUSIONS In this study, a stereotactic BBB opening system was by the Hough transform. The axons were quantified based on their extended to incorporate a feedback control algorithm based on the length marked by the tau protein. The same brain slices were utilized sub-/ultra-harmonic emissionsfrom microbubbles. The AUC responses to quantify the hippocampal density of the CD68 marker by inten- were characterized and stable feedback control was demonstrated sity-based quantification. both in-vitro and in-vivo. Using infusion injection instead ofbolus in- RESULTS Figure 1 shows two representative examples of the con- jection, combined with the real-time feedback control system, a trol and the treatment group. Following the hippocampal forma- more reliable BBB opening can be achieved across a larger brain re- tion of the control brain, it can be observed that both gion for applications focusedon drug delivery to peritumoral regions. somatodendritic and axonal tau (red) are evident. In particular, the tau marker engulfs the cell bodies and the entire in-plane length of the axons can be detected. Although the cell bodies af- fected by tau protein are also evident in the animals that re- ceived four sonications, the axonal tau was less pronounced. More specifically, the axonal distribution of the tau protein was not continuous. Differences across hemispheres were only de- tected in the treatment group. Moreover, the phagocytic micro- glia (green) seem almost absent in the control brains while they can be observed in both hemispheres of the treatment group. Quantification of the tau distribution and density are currently ongoing. CONCLUSIONS Focused ultrasound-induced blood-brain barrier opening affects tau distribution after unilateral application. Axonal tau was significantly less pronounced in the sonicated region. Whether the tau protein reduced after sonication has been trans- ferred to other cell bodies or has been entirely eliminated from the brain remains to be determined by the ongoing quantification process. Microglia were clearly activated by the sonication but its re- Fig. 1 (abstract O6). See text for description lationship to tau re-distribution remains to be established. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 5 of 122 of the discrepancies indicates that there is not any systematic error associated (mean spatialerror tends toward zero). Simulations in test cases also validated the potential of the method by showing its ac- curacy in a wide range of configurations. CONCLUSIONS A method of cavitation localization based on a three hydrophone network was explored. Localizations performed with this method were corroborated by high-speed observation in water. Also, simulations show that the method can be applied in a wide range of cases. It was shown that the use of a relatively narrow frequency bandwidth around the sub-harmonic frequency permitted to en- Fig. 1 (abstract O7). Hippocampal formation counterstained for tau hance the accuracy of the method. It is hypothesized that this comes protein with AT8 (red) and microglia activation with CD68 (green). from the fact that this signalis emitted only by oscillating bubbles The first row corresponds to the ipsilateral and contralateral side of and not by reflections of the excitation signal. In addition to the re- a control brain and their magnified regions. The second row sults shown in this study, this method could be augmented by a corresponds to the ipsilateral and contralateral side of a treated complete characterization of the cavitation event by the analysis of brain and the corresponding magnified regions the cavitation signals. Also, by adding a hydrophone, 3-dimensional localization can be performed at minimal numerical cost. The ob- tained results and the possibilities that could be explored in the fu- ture give credit to this technique as a versatile tool for O8 cavitationposition monitoring, a major step for clinical transfer in all Evaluation of a three hydrophone-based method for cavitation applications related to cavitation. localization 1 2 2 1 Maxime Lafond , Cyril Lafon , Jean-Louis Mestas , Shin-ichiro Umemura Graduate School of Biomedical Engineering - Umemura-Yoshizawa Laboratory, Tohoku University, Sendai, Miyagi, Japan; LabTAU, INSERM U1032, Université deLyon, Lyon, France Correspondence: Maxime Lafond Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O8 OBJECTIVES In the context of clinical applications, the complete characterization of the cavitation activity includes also the cavitation localization. However, the current methods for cavitation localization present various disadvantages such as the limited bandwidth of the echo imaging probes, high numerical cost and the need forexpensive equipment. This limits strongly the transfer of cavitation-based appli- cations toward their clinical use. We propose to apply a source localization algorithm to cavitation monitoring and characterization. In the context of cavitation-related therapy, it is expected to increase the reliability level cavitation applications, giving it morecredit for clinical transfer. In order to overcome these limitations, we intend to use a PVDF hydrophone network to localize the bubble cloud, con- sidered as a simple acoustic source. A classic method for source localization is triangulation. The localization of the cavitation cloud is deduced from the delays ob- tained between threereceptors with known positions. In our case, Fig. 1 (abstract O8). Comparison between highspeed camera the receptors are PVDF hydrophones. Two confocal transducers are observation and the position of the cavitation cloud calculated emitting a pulse at 1.1 MHz in order to generate cavitation in the op- with the method based on three hydrophones. While the algorithm tical field of a high-speed camera. The signals from the three hydro- based on the full frequency bandwidth (red) gives an approximate phones were recorded during the US pulse on a digital oscilloscope position of the cavitation cloud, the one based on a bandwidth and the delays between the hydrophones were calculated by finding reduced around the subharmonic frequency (blue) give an enhanced the delay maximizing the inter-correlation between the recorded sig- localization of the cavitation cloud. The green circle denotes for the nals. The source position calculated from thedelays was finally super- focal point of the US apparatus, used for the superimposition of the imposed over the images from the camera (Fig. 1). The positions calculated position over the highspeed images calculated with this method were compared to the positions of the clouds visually estimated. The mean discrepancy was calculated. The method was firstly applied using the signals with full frequency bandwidth. Then, the post-processing operation was repeated after keeping only the bandwidth of 200 kHz around the sub-harmonic frequency (550 kHz). Also, simulations are performed to evaluate the O9 versatility of the method in various test cases. Notably, spatial Design study and experimental validation of a novel phased array spreading of the source, source separation and the influence of the based on Fermat’s spiral hydrophones repartition are evaluated. 1 1 2 1 Pascal Ramaekers , Martijn de Greef , Rémi Berriet , Chrit Moonen , RESULTS The high-speed camera observations were compared with Mario Ries the localization technique for each one of 8 independent pulses. The 1 2 UMC Utrecht, Utrecht, Netherlands; Imasonic SAS, Voray-sur-l'Ognon, position of the cavitation cloudis calculated with a discrepancy of 3.1 France ±1.8 mm in the case of the full frequency bandwidth. By processing Correspondence: Pascal Ramaekers the data only in the 200 kHz frequency band around the 550kHz Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O9 sub-harmonic, the accuracy is improved to 1.4±0.8 mm. The analysis Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 6 of 122 OBJECTIVES Previous work has shown numerically that a phased extracorporeal HIFU applications the beam steering capabilities of array transducer based on Fermat’s spiral could potentially pro- the tessellated arrays are sufficient within the relevant steering vide improvements for extracorporeal HIFU therapy, in particular range. Out of the three tessellated designs, the array based on φ with respect to the energy exposure of the near field. In a first = 137.51° is technically the most favorable as it has the most step, this design study numerically evaluated phased array de- uniform element size distribution. The tapered array performs signs based onFermat’s spiral for three different spirals and two slightly better in terms of beam steering, but also induces higher different element types. The transducers were evaluated on three pressure levels in the near field. different aspects: element size distribution, beamsteering capabil- ities, and pressure distribution in the near field. In a second step, one of the designs was evaluated experimentally through hydro- phone measurements. METHODS For the numerical evaluation of the different phased array designs, the arrays were populated by 256 elements and evaluated for two different element types: circular elements of 3.8 mm diameter (sparse arrays), and element shapes generated by applying Voronoi tessellation to the predefined element po- sitions (tessellatedarrays). Three different spirals were used to determine the phased array element positions. The first spiral had a divergence angle of φ = 87.85°. Such a design has been- shown to minimize peak grating lobe levels in a study on 2D ultrasound imaging arrays based on Fermat’sspiral. Thesecond spiral was constructed using φ = 137.51° (the golden angle). Previous work has shown that using the golden angle optimizes the packing efficiency of the spiral, which ensures the most uni- form element size distribution. This has practical implications with respect to construction and electrical matching of the Vor- onoi tessellated array. The third spiral was also based on the- golden angle, but with a Taylor tapering window characterized by a sidelobe level of -30 dB applied. It has been shown for array antennas that such a tapering windowreduces sidelobe levels of the array. All transducers were designed using an aper- ture diameter and radius of curvature of 16 cm (f-number = 1). An overview of theevaluated tessellated arrays is shown in Fig. 1. Acoustic simulations were performed under free-field condi- tions using propagation of the angular spectrum of planewaves. For the experimental validation of one of the designs, pressure measurements were conducted as a function of instantaneous Fig. 1 (abstract O9). See text for description acoustic output power using a fiberoptic probe hydrophone (FOPH 2000, RP Acoustics). RESULTS Considering the element size distribution for the Voronoi tessellated arrays, the array based on φ = 137.51° was shown to be the most uniform, with all element sizes within 20% difference from the mean element size. The array based on φ = 87.85° had 6 notable outliers, but was otherwise comparable to the array basedon φ = 137.51°. The array based on φ = 137.51° and a tapering window was shown to have a highly nonuniform distribution of element sizes, with only 89 elements within 20% difference from the mean element size. Maximum pressure levels obtained for off-axis sonications are shown in Fig. 2. The sparse arrays performed comparable in terms of beam steering, with the tapered array giving a slightly better per- formance than the other two. For the tessellated arrays the tapered array gave the best off-axis performance, followed by the array base- don φ = 137.51°. The worst performance is observed for the array based on φ = 87.85°. Figure 3 shows cumulative histograms of pres- sure levels in the near field (10 < Z < 90 mm prefocal) for an on axis sonication for each of the six arrays. The sparse arrays gave a highly similar performance in terms of near field pressure distribution. The tessellated arrays substantially reduced near field pressure levels compared to thesparse arrays. Comparing the tessellated arrays, the best performance was observed for the array based on φ = 137.51. Figure 4 shows pressure as a function of time for the hydrophone measurements on one of the tessellated transducers for different in- stantaneous acoustic output powers. Rarefactional and compressional pressures ranged from 14.47 and 66.53 MPa respectively at 100 W, to 40.00 and 237.50 MPa at 800 W. CONCLUSIONS Tessellated arrays provide an improvement over sparse arrays in terms of near field energy exposure. This comes Fig. 2 (abstract O9). See text for description at the cost of reduced beam steering capabilities, but for Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 7 of 122 OBJECTIVES Even if bubble collapses are commonly thought to be the key element of cell permeabilization for drug delivery applica- tions, recent works have demonstrated the possibility of transfecting cells by gentle oscillating bubbles (stable cavitation), possibly result- ing in lower cell lysis or tissue damages. Nevertheless, in a bubble cloud, both stable and inertial cavitation activities would naturally co- exist, thus making difficult to quantify the contribution of both re- gimes on the drug deliveryprocess. In this work, we present a regulation system aiming at distinguishing the two regimes and con- trolling the stable cavitation activity during time. METHODS A feedback-loop process is implemented on the level of the subharmonic component emitted from a bubble cloud and recorded by a needle hydrophone (Fig. 1). The cloud is created by a focused transducer emitting a 550 kHz pulsed sine wave (duty cycle: 0.2; cycle duration: 250 ms; sonication duration: 60 s) in a watertank, either in the bulk (free configuration) or in a plastic tube located at the focal point (vessel-confined configuration). The feedback-loop performs a real-time modulation of the applied voltage at a 250μs loop rate, allowing to control the subharmonic emission (SC index), as well as measuring iner- tial cavitation activity (broadband noise emission, IC index). RESULTS Without regulation (open loop), the measure of the mean SC is not reproducible for a given acoustic intensity (Fig. 2-a), and the SC and IC indices oscillate during all the sonication time (Fig. 2- b). In contrast, the regulation system leads to reproducible results, showing a SC index which always reaches the feedback-looptarget value (Fig. 2-c) and remains constant during time (24 dB in Fig. 2-d). Fig. 3 (abstract O9). See text for description Moreover, as showed in Fig. 2-c, the feedback-loop allows achieving a high subharmonic response (up to 20dB) without broadband noise emission (0-5 dB), thus limiting the inertial cavitation effect. Finally, for a given high SC level, the results show a gain up to 20% of acous- tic energy in comparison to the sonication without regulation. CONCLUSIONS Evidences of control of the stable cavitation activity are reported, associated with (1) the control of the stochastic behavior of the bubble population, (2) a stable SC index during time, (3) the minimization of inertial cavitation activity, thus limiting the destructive effects due to bubble collapses, and (4) the saving of acoustic energy required for initiating stable cavitation. [Work supported by the French National Research Agency, LabEx CeLyA (ANR-10-LABX-0060) and granted bythe ANR-MOST project CARIBBBOU (ANR-15-CE19-0003)] Fig. 4 (abstract O9). See text for description O10 Real-time control of the stable cavitation activity in free and vessel-confined bubble clouds 1 1 1 Corentin Cornu , Matthieu Guedra , Jean-Christophe Bera , 2 3 1 Wen-Shiang Chen , Hao-Li Liu , Claude Inserra Univ Lyon, Université Lyon 1, INSERM, LabTAU, F-69003, LYON, France, Lyon, France; Department of Physical Medicine & Rehabilitation, National TaiwanUniversity Hospital, Taipei, Taiwan; Department of Fig. 1 (abstract O10). Schematic representation of the experimental Electrical Engineering, Chang Gung University, Taoyuan City, Taiwan invitro device with focused transducer in a water tank. Realtime Correspondence: Corentin Cornu monitoring and feedbackloop process are described Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O10 Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 8 of 122 to predetermined targets to achieve multi-position neurostimulation. According to the classical ultrasound imaging technique by array transducer, special array transducer can be used to perform dual functions of ultrasound imaging and neurostimulation. This study in- vestigates the feasibility of using linear array ultrasound transducer system to achieve image-guided dual-target ultrasound brain stimu- lation on mouse. METHODS A dual modes ultrasound system, showed as Fig. 1(A), is developed to perform B-mode imaging for predetermination of stim- ulated target and to produceacoustic radiation force for neurostimu- lation. The stimulation position was controlled by beamformer inside the field programmable gate array device. Brain imaging was done first and followed by image guided stimulation. 5 male BALB/c mice, 8-10 weeks old, 20g (+/-25%) in weight were used. Mouse was anes- thetized byintraperitoneal injection of ketamine (70mg/kg) and xyla- zine (7mg/kg) cocktail. The mouse's hair was cropped by scissor and removed by depilatory. The mouse was laid prone on an automatic Fig. 2 (abstract O10). Reproducibility experiments in pulsed sonication heating pad (69002, RWD Life Science Co.) at 37 degrees centigrade for the SC (subharmonic emission level) and IC (broadband noise level) and with its head gently immobilized using stereotaxic frame (68028, indicators as a function of the applied voltage in open loop (2-a) and the RWDLife Science Co.). Different positions were selected to evaluate SC target value in closed loop (2-c). Examples of temporal stability curves the effect. Stimulated effect was evaluated by Electromyography sig- for the evolution of the SC and IC indicators versus time at the applied nals recorded by an electrophysiological acquisition system (MedLab- voltage 180 mV in open loop (2-b) and at the SC target value 24 dB in U8C502, MedEase Ltd., Nanjing, China). The motion responses of closed loop (2-d) mouse were captured by a camera (HD1080P, AoniLtd., Shenzhen, China) in real-time. RESULTS Figure 1(B) demonstrates a B-mode image of a mouse brain for predetermining two different stimulation targets. The exact posi- tions of the targets can be selected by the computer mouse, and are pointed out by the arrows. Figure 1(C) indicates the different effects of the EMG responses evoked by the ultrasound stimuli on the target O11 A and B. Our results indicate that imaged-guided dual-target brain Research platform for rodent studies of wavefront engineered stimulation by array ultrasound can evoke different motion responses ultrasonic neuromodulation on mouse. 1 1,3 1 1 1 Steve Krupa , Eilon Hazan , Omer Naor , Michael Plaksin , Inbar Brosh , CONCLUSIONS The imaged-guided dual-target brain stimulation sys- 2 2 1 3 1 Noam Maimon , Yoav Levy , Eitan Kimmel , Itamar Kahn , Shy Shoham tem can achieve dual targets brain stimulation with varying spatial Department of Biomedical Engineering, Technion I.I.T, Haifa, Israel; and temporal conditions underthe guidance of B-mode imaging, 2 3 Insightec LTD, Tirat HaCarmel, Israel; Department of Medicine, which offers a useful tool for the applications on neural circuits stud- Technion I.I.T., Haifa, Israel ies and clinical therapies. Correspondence: Steve Krupa Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O11 This abstract is not included as it has already been published: Shoham S, Krupa S, Hazan E, Naor O, Levy Y, Maimon N, Brosh I, Kim- mel E, Kahn I. A126 Research platform for rodent studies of wave- front engineered ultrasonic neuromodulation. J Ther Ultrasound. 2016; 4(Suppl 1):31. Available from: https://jtultrasound.biomedcen- tral.com/articles/10.1186/s40349-016-0076-5 O12 Image-guided dual-target brain stimulation on mouse by array ultrasound Guofeng Li, Jiehan Hong, Qiuju Jiang, Peitian Mu, Ge Yang, Congzhi Wang, Weibao Qiu, Hairong Zheng Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences, Shenzhen, China Correspondence: Guofeng Li Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O12 Fig. 1 (abstract O12). (A) shows the block diagram of the proposed imagedguided dualtarget ultrasound brain stimulation system. Figure OBJECTIVES Ultrasonic neuromodulation has become a promising (B) demonstrates a Bmode image of a mousebrain for predetermining approach for neural science and clinical application. Currently single two different stimulation targets. The exact positions of the targets can element focused ultrasound transducer was widely used for perform- be selected by the computer mouse, and are pointed out by the ing ultrasonic stimulation. However, the stimulated position is fixed, arrows. Figure(C) indicates the different effects of the EMG responses and is hard to be precisely identified, when performing the brain evoked by the ultrasound stimuli on the target A and B stimulation. It is important to precisely steer the focus of ultrasound Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 9 of 122 O13 capabilities of the system were preserved. Among the tested gates, 2ms Pseudo-random gated sonications for reducing cavitation during period pseudo-random codes seem to be the modulation that increases thermal ablation in the brain the cavitation threshold the most while preserving heating capabilities. 2,1 3 3 3,4 Cyril Lafon , David Moore , Matthew Eames , John Snell , James Indeed, very fast switches may be associated to electronic difficulties for 2 3,4 Larner , Neal Kassell non-demonstrated benefit in terms of cavitation reduction. Work spon- 1 2 LabTAU, INSERM, Lyon, France; Department of Radiation Oncology, sored by the FUS Foundation through the Robert Merkin Fellowship and University of Virginia, Charlottesville, Virginia, USA; FUS Foundation, performed with the technical support of InSightec. Charlottesville, Virginia, USA; Department of Neurosurgery, University of Virginia, Charlottesville, Virginia, USA Correspondence: Cyril Lafon O14 Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O13 A flat array with 4096 elements for MR-guided focused ultrasound therapy 1 1 1 OBJECTIVES Transcranial HIFU is now used in clinics for treating es- Yuexi Huang , Ben Lucht, Rohan Ramdoyal , Samuel Gunaseelan , 1 1 1,2 sential tremor and proposed for many other brain disorders. This Tyler Portelli , Ping Wu , Kullervo Hynynen 1 2 promising treatment modality still faces several limitations. HIFU-in- Sunnybrook Research Institute, Toronto, Ontario, Canada; Department duced thermal ablation in the brain requires high energy resulting of Medical Biophysics, University of Toronto, Toronto, Ontario, Canada eventually in undesired cavitation and potential side effects. Original Correspondence: Ben Lucht strategies should be tested to increase treatment safety and efficacy. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O14 The goals of the present work were: 1- to evaluate the potential in- crease of the cavitation threshold using pseudo-random gated soni- OBJECTIVES MR-guided focused ultrasound has been demonstrated cations and 2- to assess the heating and steering capabilities with in various applications for non-invasive thermal ablations. The such sonications. The performance of pseudo-random sonications current clinical devices for body applications use spherically curved was compared with conventional continuous exposures. phased arrays, which improve focus quality for a limited number of METHODS The experiments were performed with the transcranial transducer elements (~200), at the expense of a limited steering MR-compatible ExAblate Neuro system (InSightec). It is a 1024-elem- range, therefore mainly rely on mechanical positioning for covering a ent, 30 cm diameter, 15 cm focal length, transducer operating at a large treatment volume. In this study, we developed a fully popu- frequency of 660 kHz. Four methods of sonication were compared: lated flat array, which allows for amuch wider steering range. Feasi- continuous wave (CW), gated emissions with pseudo-random 2ms (p- bility using this new design for thermal ablation and hyperthermia Rand2ms) or 33 us (p-Rand33us) -period codes and 30 kHz square over large target volumes was demonstrated in animal studies. (Squ30kHz). The duty cycle (DC) was set to 50% for the gated sonica- METHODS A flat array of 14 cm in diameter with 4096 elements was tions. The cavitation threshold was first evaluated in water. For each manufactured in house with center-to-center element spacing of condition, electrical driving power was increased step by step from half-wavelength at the centre frequency of approximately 500 kHz. 20 to 500W and the acoustical noise was recorded with the inte- Custom driving electronics were built using Application-Specific Inte- grated passive cavitation detectors. The spectral energy was aver- grated Circuit (ASIC) technology. The MR-compatible arrayand driving aged over the 250-500 kHz bandwidth and the sonication time (10s). system were mounted on the standard bed of an MR scanner Heating trials were then performed in a hydrogel tissue mimicking (MR750, GE Healthcare, Milwaukee, WI, USA). In vivo studies were material (TMM, ATS Laboratories). The electrical power was set to 15, performed on 16 pigs on thigh muscles. Acoustic power up to 240W 30 or 60 W, the exposureduration to 9 or 18 s. For a fair comparison over 50s were applied with MR thermometry monitoring. Focus was with CW, 50% DC sonications had either the electrical power or the either stationary during sonications or steered in circularpattern lat- exposure duration doubled. The temperature was measured by MR- erally or along the acoustic beam. The minimum delay between thermometry when focusing at the geometrical focus and when steering spots for loading new phasing parameters was 3 ms. Lesions steering the beam off-focus by 5mm-steps. were measured by T2-weighted imaging (FRFSE, TR 5000ms, TE RESULTS The assumption was made that the occurrence of cavitation 100ms). For hyperthermia applications, multi-foci sonication pattern resulted in an inflexion of the scattered energy vs. power curve. Due was applied to achieve heating over a large volume (30cm3) for 15 to frequency modulation, harmonics occur in the bandwidth of inter- min at 43°C. est for the p-Rand33us and Squ 30kHz sonications and their energy RESULTS The array focused energy well by using geometric beam- increased also with power. The threshold was then harder to assess. steering phase delays and was able to provide sustained acoustic Sudden changes in the scattered energy occurred at electrical pow- power with repeated treatments.Temperature was above 60°C at focus. ers of 180, 400, 500, 400W for CW, p-R and 2ms, p-Rand33us and Good volumes of thermal coagulation were achieved with a wide steer- Squ30kHz respectively. Unsurprisingly, higher thresholds were mea- ing range as large as 12 cm. For hyperthermia, heating around 43°C sured when repeating the experiments in more degassed water. over the large volume was observed by MR thermometry. These settings did not allow to heat the TMM when electronically CONCLUSIONS In this study we demonstrate for the first time that fully steering the focus more than 15mm-off the geometrical focus. Doub- electronically steered arrays are feasible. With a flat design and transducer ling the exposure duration did not allow to compensate the 50% DC element spacing at half-wavelength, wide steering was achieved without because of thermal diffusion. At a power of 30W, a maximal sacrificing focus quality. A wide steering range also eliminates the need temperature rise of 9°C was measured with gated sonications applied for a motorized positioning system, which simplifies system design and al- for18s while CW 30W sonication for 9 s gave a 13°C increase in lows precise MR thermometry without motion induced artifacts. temperature. However, the heating patterns obtained at 60W with p- R and 2ms and 30 W in CW were very similar for a constant 9s dur- ation. For the same conditions, lower temperature rises were mea- O15 sured for p-R and 33us and Squ30kHz. Several assumptions can be Effects of rarefactional pressure from pulsed focused ultrasound in made to explain this difference: the gating is too rapid for reaching murine melanoma and breast tumor models: implications for steady state, the amplifier struggles to switch so quickly, or cavitation immunotherapy enhanced heating is reduced. Omer Aydin, Scott R. Burks, Saejeong Kim, Joseph A. Frank CONCLUSIONS Coded sonications were proposed using rapid ran- Radiology and Imaging Sciences, NIH Clinical Center, Bethesda, dom phase shifts (Chapelon et al. 1996) or chirp (Tang and Clement Maryland, USA 2009). In the present work, it has been demonstrated on a clinical Correspondence: Omer Aydin Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O15 device that randomly gated signals also increase the cavitation threshold in water. The general idea consists in breaking the dynamic OBJECTIVES Focused ultrasound (FUS) has long been used for ther- of the bubbles moving from monochromatic to broadband sonica- mal ablation of malignant tumors. The potential role for nonthermal tions. It was then necessary to check that the heating and steering Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 10 of 122 pulsed FUS (pFUS) as amethod to modulate the innate immune re- O16 sponse to tumors is under investigation. We have previously charac- Boiling histotripsy ablation of renal carcinoma in the Eker rat terized the molecular changes in numerous tissues and have produces significant changes in the immune system 1 1 2 3 demonstrated changes in the tissue microenvironment and immune George R. Schade , Wayne Brisbane , Stella Whang , Yak-Nam Wang , 2 4 5 3 cell phenotypes following pFUS. However, the parameters needed to Kayla Gravelle , Venu Pillarisetty , W. Conrad Liles , Vera Khokhlova , 3 2 2 maximize pFUS-inducedmolecular changes in tumor microenviron- Michael Bailey , Tatiana D. Khokhlova , Joo Ha Hwang ments that could modulate immunotherapeutic response are essen- Department of Urology, University of Washington, Seattle, Washington, tially uninvestigated. We treated the B16 murine melanomaand 4T1 USA; Department of Gastroenterology, University of Washington, breast cancer flank tumors with pFUS of 1 MHz over a range of peak Seattle, Washington, USA; Center for Industrial and Medical Ultrasound, negative pressures (PNP) (1–8 MPa) without microbubble infusions University of Washington, Seattle, Washington, USA; Department of and evaluated the molecular response to sonication. Surgery, University of Washington, Seattle, Washington, USA; METHODS Subcutaneous B16 or 4T1 tumors were established in the Department of Medicine, University of Washington, Seattle, legs of C57BL/6 (B16) or BALB/c (4T1) mice and allowed to reach ~8 Washington, USA mm in diameter. Magnetic resonance-image-guided pFUS with passive Correspondence: George R. Schade cavitation detection (RK100, FUS Instruments, Toronto, ON, Canada) or Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O16 ultrasound-image-guided pFUS (VIFU, Alpinion Medical Systems, Both- ell, WA) were administered at 1 MHz over a range of PNP values (1, 2, OBJECTIVES Focused ultrasound (FUS) exposures have demonstrated 4, 6, and 8 MPa). One hundred sonications were delivered to each ras- the ability to modulate the immune system to stimulate an anti-tumor terpoint using a 10% duty cycle and a pulse repetition frequency of 10 immune response. Ourgroup has been developing boiling histotripsy Hz. During treatment planning, the entire tumor volume was covered (BH) as a non-invasive treatment for renal carcinoma (RCC) and have pre- and 2-mm spacing was used between points. Twenty-four hours after viously shown short-term changes in systemicand local cytokines and in- pFUS, tumors were harvested for histology and molecular analyses. filtration of CD8+ T-cells following BH treatment in vivo. We characterized RESULTS During pFUS, half-harmonic emissions were only detected the long-term immune response to BH RCC tumor ablation in a ratmodel. only at 8 MPa PNP in B16 tumors and at 6 and 8 MPa PNP in 4T1 tu- METHODS RCC bearing genotyped Eker rats (Tsc2 heterozygotes) were mors. Twenty-four hours after pFUS, we examined expression of randomly assigned to transcutaneous BH or FUS SHAM procedure target- intercellular adhesion molecule (ICAM), vascular cell adhesion mol- ing ~0.5 cc of RCC or non-tumor bearing normal kidney. BH was deliv- ecule (VCAM), and cyclooxygenase (COX2) in each tumortype. ICAM ered with a 1.5 MHz US-guided small animal FUS system (VIFU-2000, was elevated in 4T1 and B16 tumors between 2 and 4 MPa (P <0.05) Alpinion) operated at duty cycles of 1-2%, 10-20ms pulses, 525-600 W compared to untreated tumors (Fig. 1). We did not observe statisti- electric power. Following treatment rats were recovered, underwent serial cally significant elevations in COX2 for B16 tumors, but COX2 was el- US imaging surveillance and serial blood draws, and were survived for7, evated in 4T1 tumors treated with pFUS as 4, 6, and 8 MPa (P <0.05). 14, or 56 days. Following euthanasia, the treated and contralateral kidney, VCAM expression was not upregulated bypFUS in either tumor type tumor draining lymph nodes (TDLN), and spleen were collected. Flow-cy- at any pressure. Furthermore, H&E staining revealed wide-spread co- tometry was performed on processed tissues and blood to analyze for agulative necrosis and red blood cell extravasation in both tumor changes in circulating and local immune cell populations. typestreated at 8 MPa compared to lower pressures. RESULTS BH treatment was associated with significant alterations to the CONCLUSIONS We found that pFUS could induce molecular changes immune system within 2 weeks vs. SHAM treatment with characteristics of in B16 and 4T1 tumors that are consistent with molecular response an anti-tumor immune response (see Fig. 1). Specifically, BH was associated that can influence immune cell tropism. Interestingly, we demon- with a significant 3.4-fold (p=0.048) increase in splenic CD11c+ antigen pre- strate that there may be optimal rarefaction pressures to induce ap- senting dendritic cells. BH also produced alterations in CD4+ helper T-cell propriate molecular changes in tumors. For example, the greatest populations with a significant 1.2-fold (0.041) increase in CD4+ CD62L- increases in ICAM, that could facilitate immune cell infiltration, was CD44+ effector memory cells and near significant 1.8-fold (p=0.08) increase observed below the cavitation threshold in both tumor types. Fur- in central memory CD4+ CD62L+ CD44- cells in TDLNs. Finally, BH treat- thermore, we show that molecular responses can be fundamentally ment resulted in significant alterations in cytotoxicCD8+ T-cell populations. different across tumor types (e.g. changes in COX2 in 4T1 tumors, Specifically, BH treatment was associated with significantly increased CD8+ but not B16 tumors), that may reflect differences in the physiological CD62L- CD44+ effector memory cells in both TDLNs and spleen (3.0-fold, and functional responses. These results suggest it may be possible to p<0.01 and 2.9-fold, p=0.03, respectively) and central memory CD8+ CD62L use pFUS to modulate immune function, but that a greater under- + CD44- cells in TDLNs (6.8-fold, p <0.01) with a corresponding significant standing of pFUS molecular effectsis needed. Given the heteroge- decrease in CD8+ CD62L+ CD44+ naïve cells in TDLNs (0.4-fold, p=0.01). neous response across tumor types, that each pFUS parameters may CONCLUSIONS These data represent the first extensive analysis of the im- need to be empirically evaluated. mune response to BH tumor treatments and suggest that BH RCC ablation produces significant changes to the immune system suggestive of an anti- tumor response. Analysis of longer-term survival (56 days) is ongoing and will demonstrate if these changes are long-lived. Future studies will further evaluate the specificity of this response and whether it can improve clinic- ally relevant outcomes. Funding: The Focused Ultrasound Foundation, The Urology Care Foundation, and NIH K01EB015745 and R01CA154451. Fig. 1 (abstract O16). Flow cytometry results from tumor draining lymph nodes and spleens 2 weeks following SHAM or BH treatment Fig. 1 (abstract O15). See text for description of RCC Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 11 of 122 O17 CONCLUSIONS The results demonstrate that USMB treatments in Potentiating checkpoint inhibitor therapy with ultrasound combination with anti PD-1 therapy resulted in a significant inhib- stimulated microbubbles ition of tumor growth. The data suggest that USMB therapy may be 2 1, 2 1, 2 1, 2 Sharshi Bulner , Aaron Prodeus , Jean Gariepy , Kullervo Hynynen , potentiating anti PD-1 therapy through a T-cell dependent mechan- 1, 2 David Goertz ism. This approach is a means by which to enhance checkpoint in- 1 2 University of Toronto, Toronto, Ontario, Canada; Sunny brook Research hibitor effects without the deliterious side effects associated with Institute, Toronto, Ontario, Canada adding other immunotherapy or chemotherapy agents. Correspondence: Sharshi Bulner Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O17 O18 OBJECTIVES Immunotherapies are playing an increasingly prominent Proteomic and histological effects of pulsed focused ultrasound in role in the treatment of cancer, driven in large part by the success of the rat heart 1 1 1 1 checkpoint pathway inhibitors such as anti-PD-1 monoclonal anti- Kee W. Jang , Tsang-Wei Tu , Scott R. Burks , Bobbie K. Lewis , Joseph A. 1,2 bodies. These approaches inhibit the ability of tumors to evade the Frank immune system and boost the activity of anti-tumor Tcells. However, Radiology and Imaging Sciences, National Institutes of Health, Bethesda, potent and durable results are to date only obtained in subsets of Maryland, USA; National Institute of Biomedical Imaging and patients with certain cancers (e.g. melanoma). They are less effective Bioengineering, National Institutes of Health, Bethesda, Maryland, USA in tumors with low immunogenicity, and those with low levels of Correspondence: Kee W. Jang tumor infiltrating lymphocytes (TILs). As a result, it is increasingly rec- Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O18 ognized that combination therapies will be required. While it is known that different forms of therapeutic ultrasound can elicit a OBJECTIVES The purpose of this study was to investigate the effects spectrum of immune responses, it remains to be established if they of pulsed focused ultrasound (pFUS) exposures to the rat heart and can be harnessed to potentiate the effects of checkpoint inhibitors. evaluate the proteomic and histological changes temporally follow- In this study we examine the combination of anti-PD-1 therapy with ing sonication. ultrasound stimulated microbubbles (USMBs). Inparticular, we use a METHODS Eight to ten week old female Sprague Dawley rats were level of ultrasound exposure that is non-thermal, but elicits a signifi- imaged on a 3T MR scanner (Achieva, Philips Healthcare, USA) and cant degree of vascular injury. T2-weighted MR images were acquired with 8.9ms repetition time METHODS A murine colorectal cancer cell line (CT26.wt) was (TR), 4.5ms echo time (TE) in 1mm slice thickness through the chest employed to initate tumors subcutaneously in the hind limbs of fe- wall in sagittal plane. The MR images were used as guidance for male Balb/c mice. There were four study groups: microbubbles-only pFUS to target the left ventricular apical myocardium (center fre- (control/sham treatments), Anti PD-1 (drug-only group), USMBs (ultra- quency 1.1MHz; PNP 3MPa; 10ms bursts; 1Hz PRF; 100 sonications/ sound-only treatment group) and USMBs + Anti PD-1(combined point, 40 points to the apex, RK100, FUS Instruments, CAN). Follow- treatment group). Experiments commenced when tumors were well ing pFUS treatment, proteomic analysis was performed of sonicated established, in the range of 50-100 mm3. The immunotherapy drug myocardium lysate (n=5/timepoint) over 24 hrs along withcardiac in- used in this study is ananti-mouse PD-1 (clone: RMP1-4, Bioxcell) and jury markers from the myocardium and serum (s) using single- or was administered at a dosage of 200 μg intraperitoneally prior to multi- plexed ELISA. In order to monitor and examine the activity of ultrasound or sham treatments, and then subsequentlyadministered cardiac function, electrocardiogram (EKG) was recorded and analyzed every 3 days for a total of 5 doses. The microbubbles employed were at pre-determined time points using EKG recording module (IX- an experimental phospholipid encapsulated agent administered in ECG12, iWorx Systems, Inc., NH). pFUS treatedmyocardium (n=3) bolus form prior toinsonation. Ultrasound exposures were delivered were stained with haematoxylin and eosin (H&E) to evaluate the with a single element focused 1 MHz transducer with the entire tu- morphology. Statistical analysis was performed using ANOVA with mors within the -3dB beam width. The pulsing scheme was to send multiple comparisons to sham control with p<0.05. 50 0.1 ms long pulses (1.6 MPa peak negative pressure) spaced 1 ms RESULTS pFUS treatment to the myocardial apex induced a delayed apart, which was repeated at 10 second intervals for a duration of 3 pro-inflammatory cytokine expression with significant elevations in minutes. This low duty cycle approach avoided thermal elevations interleukin (IL) 1α,1β, 17tumor necrosis factor alpha (TNFα) inter- and has been previously established to induce vascular injury in tu- feron gamma (IFNγ) and anti-inflammatory factors including IL4, 10 mors. Antitumour effects was assessed withlongitudinal experiments along with chemo-attractant factor regulated on activation, normal T (n=5-7 per group), where tumor growth was evaluated every three cell expressed and secreted (RANTES). Significant (p<0.05) cardiac days until ethical size endpoints were reached. Two sets of acute ex- biomarkers including cardiac troponin (cTn) and N-terminal pro b- periments were conducted for all groups, where tissue (tumors, type natri-uretic peptide(NT-proBNP) were also detected in both spleens and tumor draining lymph nodes) was harvested at either serum and myocardium that would be indicative of cardiac injury 3 or 7 day time points post induction (4-6 pergroup). Tumor tis- (Fig. 1). The expression of proinflammatory cytokines including be- sue underwent flow cytomtery analysis to assess immune cell sta- came significant (p<0.05) at 12 through 24 hrs after pFUS treatment. tus. Splenic and lymphatic tissue was subject to cell count The increase in cyclo oxygenase 2 (COX2) at 24 hours would suggest analysis and ELISPOT to quantify the expression of IFN-gamma as that the molecular changes would be occurring through NFkB path- a metric of T-cell activation. way. The significantly increased levels of cardiac injury markers, NT- RESULTS It was observed that the combined treatment group had proBNP and cTnI, in both myocardium and serum were observed a significantly smaller tumor volume compared to the sham after 18 hrs following pFUS treatment. Analysis of EKG signals re- group, USMB-only group and anti PD-1group at Day 6 (p<0.001, vealed that the heart rate declined immediately after pFUS and p<0.05 and p<0.05, respectively) and at Day 9 (p<0.001, p<0.001, within 1 hour returned back to normal (Fig. 2). H&E stains did not p<0.001, respectively). Using a size dependant endpoint (>1000 show morphologic changes in the apex in pFUS treated myocardium mm3), the combined treatment group resulted in a significantly compared to sham controls (Fig. 3). longer survival time compared to MBs (p<0.01), USMB (p<0.05) CONCLUSIONS We demonstrated a reproducible and delayed mild and anti PD-1 (p<0.01). Median survival between the combined cardiac injury model using MR-guided pFUS the rat. Proteomic treatment group and individual treatments was 16 days versus 10 changes and cardiac markers wouldindicate that the evolution of in- days. Flow cytometry showed that the USMB treatments pro- flammatory response to low pressure pFUS (3MPa) starting approxi- duced elevated levels oftumor infiltrating leukocytes (TILs - CD45 mately 6 hours following pFUS would suggest that sonication is +) and that the ratio of cytotoxic T-cells (CD8+) to regulatory T causing a delayed cardiac injury. Further histological and transcrip- cells (CD45+ Foxp3+) was increased. At Day 3, the combined tomic analysis will be needed to understand the pathophysiological therapy group had significantly higher TIL and cytotoxic T-cell effects of pFUS to the heart and whether the model can be used as levels than the anti PD-1 monotherapy (p<0.05) group. a basis to evaluate myocardial contusion. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 12 of 122 OBJECTIVES Biomedical Capacitive Micro-machined Ultrasound Transducers (CMUTs) have been mainly developed for imaging pur- poses. CMUTs exhibit several advantages, such as ease of miniaturization (cell size: micron size), inherently broad bandwidth (>several MHz) and good MR compatibility. This technology also has potential for generating high intensity ultrasound, which could have an interest in developing ultrasound thermal ablation therapies. To date however, the feasibility for generating high intensity ultrasound with CMUTs has only been described in a few modeling and in-vitro studies. The use of CMUTs in continuous wave (CW) mode ofopera- tion remains challenging since it requires the development of robust cell structures and dedicated driving strategies. Here, we present an Fig. 1 (abstract O18). Proteomic analysis of pFUS treatment in a in-vivo study investigating the feasibility of generating directional rat myocardium. Pro and anti inflammatory cytokines including ultrasound-induced heating and thermal damage in brain tissue, with cardiac injury biomarkers showed delayed enhancement CMUTs designed for interstitial high intensity contact ultrasound following pFUS treatment and the increased expression lasted (HICU) applications under magnetic resonance (MR) guidance. the end of predetermined time points. Significance was METHODS Two types of CMUT prototypes were fabricated using a determined by ANOVA p <0.05 series production batch of CMUTs and a wafer bonding based process. A first intermediary CMUTprototype was made of 5 columns of 4-element 1D-linear arrays mounted on a 16 cm long, 5 mm wide rigid PCB. The elements were electrically coupled 2 by 2 within agi- ven column (element size: 2.7 mm x 0.8 mm). A final prototype was a multi-faceted CMUT-based HICU catheter incorporating ten 1D-lin- ear arrays, 32.4 mm long and 0.8 mm wide. The arrays were mounted on a cylindrical 9-French flexible catheters (20 cm long), which formed a prism-shaped 2D array for multidirectional radial ultrasound exposures. CMUT prototypes were used in a porcine model for gen- erating thermal ablations in brain tissue interstitially. Preliminary nu- merical simulations allowed identifying a range of surface ultrasound intensities (Iac) suitable for inducing thermal ablation in brain tissues Fig. 2 (abstract O18). Analysis of EKG before and after pFUS. (A) with these CMUT designs (Iac > 10 W•cm-2). CMUTs were used in CW Heart rate (HR) was immediately declined following pFUS and mode (HICU sequence: 4s ON/1s OFF, f = 7.9 MHz), and the bias and returned back to normal shortly. (B) Transient elevation of RR interval driving voltages were chosen in order to operate in the collapses confirmed the changes of HR following pFUS treatment. Significance napback regime and reach the intensity level required for thermal was determined by ANOVA ****p<0.0001 HICU lesioning. Ultrasound exposures conditions were applied through an escalation dose process (total exposure time: from 3 to 15 min, up to 10 active CMUT elements). HICU-induced thermal heat- ing generated with CMUTs was evaluated in vivo on 10 pigs and monitored under real-time multi-planar magnetic resonance therm- ometry (MRT) (Fig. 1). RESULTS Overall, directional HICU-induced temperature increases were generated in the in-vivo porcine brain with CMUTs. The heating pattern could be monitoredwith excellent time-space resolution be- Fig. 3 (abstract O18). H&E stain of pFUS treated myocardium. yond a radius of 1 mm around the CMUT device (12 MRT maps every Compared to SHAM (A), There was not difference in myocardial 1s, temperature standard deviation: ± 2.5°C). Heating patterns ex- morphology different at post 6 hrs (B), post 24 hrs (C)and post 120 tended over 1 cm from the CMUT elements within 2 min exposures. hrs (D) following pFUS. Bar = 100μm HICU exposures could be performed continuously without a water- circulating coolingsystem. After 6 min, the temperature of the brain tissue was increased locally above the 55°C threshold necessary for the creation of irreversible thermal damage (△Tmax> 35 °C). Treat- ment volumes > 1.5 cm3 could be completed within 13 min. Just after treatment, tissue changes were visible on T1- and T2-weighted anatomical images. Tissue ablation boundaries detected on T1w and O19 T2w images, respectively hypo and hyper signal boundaries, corre- CMUT transducers for interstitial MR-guided high intensity lated well with the 55°C isotherm boundaries (MRTmaps). Several ultrasound heating: in-vivo feasibility in a pig brain model 1,2 1,2 1,2 contrasts were observable on MR images within the lesion area, William Apoutou N'Djin , Jeremy Vion Loïc DAUNIZEAU , 2 1, 3 4 which were consistent with previous studies reporting the presence Christopher Bawiec , Guillaume BOUCHOUX , Nicolas Sénégond , 1,2 3,5 of coagulation andliquefaction necrosis in brain after high intensity Jean-Yves Chapelon , Alexandre CARPENTIER 1 2 ultrasound exposures. Gross sample and histological analyses con- LabTAU, INSERM, U1032, Lyon, Rhone-Alpes, France; Univ Lyon, firmed the presence of brain tissue coagulations. Université Lyon 1, Lyon, F-69003, France; CarThera Research Team, Brain and Spine Institute, Paris, France; Vermon SA, Tours, France; CONCLUSIONS The feasibility of HICU therapy with CMUTs has been Department of Neurosurgery, Assistance Publique, Hopitaux de Paris, established in vivo. Further investigations are ongoing to improve the Pitie Salpetriere, Paris, France robustness of the CMUTdevices and increase the treatment volumes. Correspondence: William Apoutou N'Djin This project was supported by CarThera, the French National Research Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O19 Agency (ANR, 2010) and Single InterministerialFund (FUI, 2013). Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 13 of 122 reconstruction algorithm implemented in a ultrasound beamformers with digital parallel processing units; 4) generating the HIFU beam image. The experimental set-up included a mono-element HIFU spherical transducer (emitter, focal distance: 90 mm, working fre- quency: 2.4 MHz) and a commercial ultrasound imaging probe (re- ceiver, model: L7-4, ATL-Philips) for collecting the emitted HIFU signals. In this experiment, the emitter was aligned with the receiver face to face each other. For the wave field reconstruction, an ultrafast ultrasound scanner (Vantage 256, Verasonics) was used for the signal recording and processing via plane-wave beamforming. Evaluations: The feasibility of the method was evaluated using either the time-re- versal reconstruction or the beamforming reconstruction, with and without HIFU beam aberrations. First, in the experiment in water without HIFU beam aberrations, performance of the method was demonstrated in comparison with the reference field obtained by nu- merical calculation of the forward propagation using the Rayleigh in- Fig. 1 (abstract O19). In-vivo interstitial HICU heating induced in a tegral and hydrophone scanning. Then, in the experiment with HIFU porcine brain with CMUTs under MR temperature monitoring beam aberrations induced by heterogeneous hydrogel, HIFU beam visibility was evaluated with referred to the HIFU pressure field mea- sured by hydrophone scanning. RESULTS In a lossless medium (water), qualitative agreements were found between the amplitude of the reference HIFU field from numerical calculation (Fig. 1a) andthe HIFU fields obtained O20 from real experimental acquisitions after time-reversal and Real-time HIFU beam imaging using beamforming in an beamforming reconstructions (Fig. 1b and c). In the case of a ultrasound scanner heterogeneous medium including multiple bubbles and cracks, 1 2, 4 3 Kazuhiro Matsui , Françoise CHAVRIER , Takashi Azuma , Ichiro the HIFU beam was aberrated and split by the ultrasound 1, 5 2, 4 2,4 Sakuma , William Apoutou N'Djin , Rémi Souchon propagation through the aberrators (Fig. 2). The aberrated HIFU 1 2 Engineering, The University of Tokyo, Tokyo, Japan; LabTAU, INSERM fields obtained by time-reversal reconstruction (Fig. 2a and d) unité 1032, Lyon, France; Medicine, The University of Tokyo, Tokyo, and using the beamforming reconstruction (Fig. 2b and e), 4 5 Japan; Univ Lyon, Université Lyon 1, Lyon, France; Medical Device closely matches the reference field obtainedby hydrophone Development and Regulation Research Center, Tokyo, Japan measurement (Fig. 2c). Correspondence: Kazuhiro Matsui CONCLUSIONS The results indicated the feasibility of the method, al- Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O20 though reconstruction errors were observed, arising from the finite aperture size of the probe or from large path lengths from the OBJECTIVES High Intensity Focused Ultrasound (HIFU) therapies can source. Despite this limitation, the proposed method may be accept- currently be assisted by medical imaging for guidance or monitoring able as an easy, simple, and real-time HIFU beam imaging techni- of some tissue characteristics (e.g. anatomical structures, physio- queused for therapy guidance if the HIFU emitter and its receiver logical parameters), but no visualization of the HIFU beam itself is can be located in a proper face-to-face orientation. currently provided for assisting the treatment targeting. In biological tissues, however, the HIFU beam may be strongly attenuated, or may even focus outside the desired target region, for example due to un- controllable experimental parameters (acoustic coupling at the inter- face of the ultrasound source, anatomical barriers such as bones, gas), or due to multiple tissue layers having different ultrasoundpro- pagation properties (e.g. muscle, skin or fat). These experimental pa- rameters are difficult to quantify precisely a priori, and cannot be completely accounted for during calculation of the focusing parame- ters.There is therefore a risk of applying the HIFU energy and thus creating a therapeutic effect (e.g. thermal lesion) at a wrong position inside the medium, which potential deleterious consequences on the efficacy, accuracy, and safety of the treatment. The object of this study is to provide a method for imaging, in real time, the HIFU beam inside an acoustically propagative medium using beamforming in an ultrasound scanner.Novelty and advanta- gesThe novelty of the method can be found in its implementation of beamforming in an ultrasound scanner, which is analogous to the backward reconstruction using time reversal. The advantage of this method is its real-time imaging capability owing to digital parallel processing of the scanner. Further advantage is simplicity of the im- aging system without requiring additional equipment aside from an Fig. 1 (abstract O20). HIFU amplitude field in a lossless medium ultrasound scanner and an ultrasound array serving as a time-reversal (water). The horizontal and vertical axes are respectively x and zaxes. mirror. Each image is normalized by maximum amplitude. (a)Numerical METHODS Imaging technique:The method for imaging in real time calculation via forward propagation using Rayleigh integral. (b) the HIFU beam comprising steps of: 1) emitting a HIFU beam in the Experimental acquisition using backward reconstruction using tissues; 2) measuring voltage signal waveform values at a reception Rayleigh integral. (c)Experimental acquisition using backward region using an ultrasound array probe serving as a time-reversal reconstruction using beamforming mirror; 3) calculating acoustic field values in the tissues using the Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 14 of 122 RESULTS MEUS treatment for 5 min successfully shut down or re- duced the blood perfusion of rabbit livers. The average peak inten- sity (PI) of CEUS dropped significantly from 84.0±4.3% to 49.3±5.1%. The average volume of ablation zone (4.55±0.83 cm3) treated by MEUS plus RFA combination was significantly bigger than that of RFA treatment alone (1.63±0.29 cm3) (Fig. 1). The levels of serum ALT and AST did rise after both of two treatments but recovered to normal eight days later (Figs. 2, 3). CONCLUSIONS Liver RFA can be greatly enhanced by combining MEUS pretreatment. Fig. 2 (abstract O20). HIFU amplitude field with aberrations obtained experimentally. Each image is normalized by maximum amplitude. (a) Wideview image obtained by backward reconstruction using Rayleigh integral. (b) Wideview image obtained by backward reconstruction using beamforming. (c) Reference image obtained by hydrophone measurements in the ROI of 8 × 20 mm2. (d) Reconstructed image via Fig. 1 (abstract O22). See text for description Rayleigh integral in the ROI. (e) Reconstructed image via beamforming in the ROI O21 Low-intensity ultrasound extends lifetimes of transplanted mesenchymal stromal cells in murine muscle Scott R. Burks, Matthew Nagle, Saejeong Kim, Joseph A. Frank Radiology and Imaging Sciences, NIH Clinical Center, Bethesda, Maryland, USA Fig. 2 (abstract O22). See text for description Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O21 Correspondence: Scott R. Burks Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O21 This abstract is not included as it has already been published: Burks S, Nagle M, Kim S, Milo B, Frank J. A90 Low-intensity ultrasound prolongs lifetimes of transplanted mesenchymal stem cells. J Ther Ultrasound. 2016; 4(Suppl 1):31. Available from: https://jtultrasound.- biomedcentral.com/articles/10.1186/s40349-016-0076-5 O22 Fig. 3 (abstract O22). See text for description Impact of microbubble-enhanced radiofrequency ablation of rabbit liver Zhong Chen, Xueyan Qiao Department of Ultrasound, Xinqiao Hospital, The Third Military Medical University, Chongqing, China Correspondence: Zhong Chen O23 Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O22 In vitro and in vivo investigation of high intensity focused ultrasound (HIFU) hat-type ablation mode OBJECTIVES To investigate the synergistic effect of combining micro- Hongya Dai bubble-enhanced ultrasound (MEUS) and radiofrequency ablation Biomedical Engineering, Chongqing Medical University, ChongQing, (RFA) on normal rabbit liver. ChongQing, China METHODS Eighteen surgically exposed rabbit livers were treated Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O23 with MEUS immediately followed by RFA. The other 18 livers were treated with RFA alone. The therapeutic ultrasound of MEUS was op- This abstract is not included as it has already been published: erated at a pressure amplitude of 4.3 MPa and a duty cycle of 0.22%. Dai H, Chen F, Yan S, Ding X, Ma D, Wen J, Xu D, Zou X. In Vitro and Contrast-enhanced ultrasound (CEUS) was used to evaluatethe liver In Vivo Investigation of High-Intensity Focused Ultrasound (HIFU) circulation. Twenty livers were removed for volume measurement of Hat-Type Ablation Mode. Med Sci Monit. 2017; 23:3373-3382. Avail- ablation and the rest livers were also harvested for histological exam- able from: https://www.medscimonit.com/abstract/index/idArt/ ination 24h after treatment. Serum ALT and AST levels of 10 rabbits 902528 DOI: 10.12659/MSM.902528 were examined to monitor any changes of liver function. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 15 of 122 O24 The skin interface was identified in imaging to be approximately Image-based predicition of focusing gain in situ using dual-mode 34mm. Based on the acoustic power delivered and using re- ultrasound arrays ported numbers of acoustic properties of muscle the calculated Brogan T. McWilliams, Dalong Liu, Emad S. Ebbini heating rate was 9629°C/s. This is consistent with imaging data Electrical Engineering, University of Minnesota, Minneapolis, Minnesota, of the delivered therapy exhibiting a bubbles from boiling condi- USA tions within <100 ms. Correspondence: Brogan T. McWilliams CONCLUSIONS An image-based method for real-time estimation of Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O24 the focusing gain using DMUA imaging data was validated in vitro at multiple frequencies. The method has been applied for the estima- OBJECTIVES We have previously described a dual-mode ultrasound array tion of the focusing gain in situ from real-time DMUA imaging data (DMUA) system for therapeutic image-guided focused ultrasound (IgFUS) in vivo. As described, STF imaging of the phantom slab allowed for applications. This system has been shown to provide closed-loop spatio- measuring the beam dimensions and the FUS interaction with the temporal control of FUS beams with 2 millisecond and submillimeter time tissue and could be used in future studies to extract details of an in- and spatial resolutions, respectively. With appropriate feedback (e.g. homogeneous medium to provide accurate estimates of the focusing temperature) this system offers the promise of delivering precision HIFU gain (or heating rate). Further validation of the calculated heating therapy, which would improve the safety and efficacy of IgFUS proce- rate will be performed in vitro and in vivo by measuringtemperature dures. A key performance parameter for HIFU therapy is the focusing gain, with thermocouples in the vicinity of the focus. which is a characteristic of the transducer (array). The imaging capability of the DMUA allow for the estimation and characterization of the precise imagingpathof the HIFU beam throughthe bolusand interveningtissue to HIFU target. The objective of this paper is to validate an image-based approach for the estimation of the heating rate in situ utilizing simplified propagation models and calibration power measurements. METHODS In this study, we used a 3.5-MHz, 64-element DMUA proto- type with concave aperture (40-mm radius of curvature with f# = 1). This prototype was used for generating the HIFU beam as well as col- lecting the corresponding imaging data. Figure 1 shows the in vitro setup used to provide a validation for the image-based approach. In particular, an acoustic power meter (Omega, Ohmic instruments Easton, MD) was modified to allow the measurement of the insertion loss due to a slab of tissue-mimicking phantom between the DMUA and the tip of the cone (treated as the target). The TM phantom was fabricated from animal skin bovine gelatine, graphite,1-propanol, glutaraldehyde, and deionized water. Absorption of the phantom is predominately due Fig. 1 (abstract O24). a) shows a digital photo of the power meter to the presence of graphite and was determined to be 0.6 and 1.0dB/ with the 3D printed support bracket and slot placed into the water cm/MHz for two 4-mm disk-shaped slaps. Two modes of the imaging tank. b) shows a cross section along the center toshow orientation were used to characterize the FUS beam propagation through the tis- of the DMUA and phantom with respect to the cone. the length d is sue: 1) Synthetic-aperture(SA) imaging, which provides larger field of adjustable for disk1 and 2 this length was set to 4 mm view (FoV) to characterize the propagation medium, and 2) Single- transmit focus (STF), which provides specific feedback about the inter- actions between the FUS beam and the tissue in its path to the target. The STF imaging is performed using the same beamforming parame- ters of the intended therapeutic HIFU beam, but at diagnostic levels and with sub-microsecond pulse duration. HIFU was applied at 4 differ- ent frequencies (2.4 to 4.2 MHz in stepsof 0.6 MHz). HIFU shots of 1-sec durations were used and repeated 4 times. SA and STF images were collected before, during and after the application of therapeutic HIFU. The STF frame rate was 400 fps, which was helpful to fully characterize the incidence of cavitation and/or instability in the power measurement. RESULTS Figure 2 shows SA and STF images with and without the phantom slab in place. From Fig. 2.b an approximated dimesion of the beam at the geometric focus was determined by the -6 dB con- tour to be approximately 0.3 x 0.5 mm. Furthermore, using a phan- tom that results in speckle reflections shown in Fig. 2.d) the beam dimensions can be observed as it traverses through the phantom. Figure 3.c) shows the thickness of the phantom to be about 4mm and ending at approximately 39mm. Figure 3 shows the average (n=4) precentage of power absorbed by the respective phantom disks and validates the approach described in methods at variousfre- Fig. 2 (abstract O24). (a) SA image of the power meter without the quencies. The calculated and measured power absorption show simi- phantom slab. (b) STF image of the power meter without the lar trends with respect to frequency. Figure 4 shows an SA image of phantom slab. (c) SA image of the power meter with the phantom a rat's left carotid artery in vivofrom an image-guided targeting ex- slab. (d) STF image of the power meter with the phantom slab periment performed using the DMUA prototype described above. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 16 of 122 the gaps between adjacent lesions may have a seeding effect, leading to further growth of malignancies due to the untreated sections. To avoid lesion interactions, several authors suggested a pre-defined time delay betweenlesions or change in exposure parameters. The former results in unnecessary treatment delays while the latter involves cap- acious real-time computations for optimizing the treatment (dynamic- ally computing thermal dose) as well as remotely and frequently switching on/off high power equipment. In order to overcome these deficiencies, we propose a method for determining the optimal lesion arrangement within any arbitrary tumor size and shape, based on an extension of the bubble packing algorithm first described by Shimada in 1995 [1]. The original algorithm was extended to allow lesions to take any arbitrary position andorientation within the specified tumor volume, and evaluated on spherical and ellipsoidal tumor models. METHODS The bubble packing algorithm first described by Shimada was extended to the case of arbitrary tumor position and orientation. The existing approach utilizes a model in which overlapping lesions apply forces to adjacent lesions until an equilibrium is reached. Additional forces are applied at the boundary of the tumor to ensure lesions remain within the tumor volume. In this work, the model has been extended such that Fig. 3 (abstract O24). Percentage of power in therapy shot absorbed in addition to the force, a torque is applied resulting in changes in lesion by phantom slab. Disk 1 and 2 corresponds to 0.6 and 1.0 dB/cmMHz orientation and allowing the optimization of lesion configuration in all six respectively and are both a thickness of 4mm dimensions. Initial evaluation of the technique was performed based on two simplified tumor models. While calculating and fitting the lesions, we assumed our HIFU probe focus as a3×7×3mm ellipsoidal volume and a 20mm diameter spherical target tumor volume, and a 10×5×5 mm ellips- oidal volume. As lesion placement may be constrained to a single acous- tic window located some distance from the organ, the case of fixed lesion orientation was also considered as well as the free orientation and trad- itional raster approach. During raster lesion placement, lesions were gen- erated at evenly spaced intervals throughout the tumor and surrounding volume; those lesions of which the centroid fell outside the tumor volume were then excluded. To allow direct comparison of the approaches, the number of lesions in the optimized cases was set to the same as that uti- lized for the raster case. In each case Monte-Carlo were utilized to esti- mate the total portion of tumor tissue destroyed, as well as the total volume overlap and volume of healthy tissue (i.e. that outside of the de- fined tumor region) ablated. Points were randomly generated within a do- main covering twice the tumor volume, with those falling within the targeted region but none of the lesion sites considered as untreated, those within both the tumor volume and one (or more) lesion volume considered as treated, while those outside of the tumor volume but within one or more of the lesions considered as ablated healthy tissue. The optimization process was completed 10 times for each configuration. RESULTS The modified bubble packing approach was successfully imple- mented in C++ and all described evaluation cases successfully performed. The achieved results demonstrate that the introduction of the bubble Fig. 4 (abstract O24). SA imaging of therapy plane with carotid packing approach leads to an improvement of approximately 10% in total body in vivo treated volume as compared to the raster approach, as well as a decrease in ablated healthy tissue of up to 20% (Fig. 1, Table 1). Slight differences in the free and fixed orientation cases, specifically increases in overlap and decreases in ablation of healthy tissue were observed with the intro- O25 duction of orientation into the optimization framework. An automatic approach to lesion planning for robotic HIFU CONCLUSIONS This work described the development and evaluation of a Tom Williamson, Scott Everitt, Ranjaka De Mel, Sunita Chauhan general automated method for the treatment planning of lesion positions dur- Mechanical and Aerospace Engineering, Monash University, Melbourne, ing HIFU. Theapproach allows the determination of optimal lesion locations Victoria, Australia and orientations, and can be applied to arbitrary tumor shapes and sizes as Correspondence: Tom Williamson well as arbitrary lesion configurations. Evaluation of the approach on spherical Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O25 and ellipsoidal tumor models demonstrated the feasibility of the approach. OBJECTIVES High intensity focused ultrasound (HIFU) based ablation has been found to be predictable and successful for treatment of ‘car- cinoma in situ’ in variousorgans. At this stage, the neoplasm is generally spherical in shape while, conventionally, focused ultrasound (FUS) treat- ment involves ‘raster’ scanning the formation ofthe HIFU lesions in the ROI, an approach that will generally not conform to the spherical tumor geometry. This may lead to two major undesirable effects: large gaps or overlaps at the tumor margins (physical spacing isn’t optimized) and between individual lesions, or significant ‘lesion-to-lesion’ interaction creating uncertainty in the shape, size and extent of the subsequent le- Fig. 1 (abstract O25). See text for description sions due to the remnant effect of previously laid lesions. Furthermore, Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 17 of 122 Table 1 (abstract O25). Summarized results of algorithm evaluation of 3.8 ± 1.5 mm. Ultrasound images revealed a hyperechoic region that was well correlated with the macroscopic analyses of the HIFU le- sions. Necrosis of placental tissues exposed to HIFU was confirmed with macroscopic and microscopic analyses. No significant variation in ma- ternal and fetalparameters was observed during HIFU exposure. Histo- logical examination demonstrated coagulative necrosis within the core of the lesion. Terminal villi were split with trophoblast desquamation, mesenchymal involution and fibrinoid deposits containing fragmented red blood cells. Congestion and hemorrhagic villitis were seen at the- border of the lesion. No damage was seen in the chorionic vessels wall or endothelial cells. There was no inflammation. CONCLUSIONS This study demonstrates in a monkey model of preg- nancy the feasibility of HIFU treatment applied to the placenta for potential application to treattwin-to-twin transfusion syndrome. We report here the first use of a completely non-invasive treatment of the placenta in pregnant animals. The primary aims of this study O26 were to assess the efficacy and safety of this technique for the preg- Extracorporeal high-intensity focused ultrasound treatment of the nant monkey and fetus. The presence of well-defined and controlled placental unit: in vivo study using a monkey model of pregnancy lesions in the placenta without complications was encouraging. Five 1 1 1 2 David Melodelima , Jonathan Caloone , Anthony Kocot , Cyril Huissoud HIFU lesions were created in the anterior placenta, and one was cre- 1 2 LabTAU - U1032, INSERM, Lyon, Rhône Alpes, France; CHU Croix ated in the posterior placenta. The insonification of the posterior pla- Rousse, LYON, France centa was possible without any complications, but the risk to the Correspondence: David Melodelima fetus is much more important. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O26 OBJECTIVES Fetal surgery represents an interesting approach for sev- eral fetal diseases, particularly for the treatment of twin-to-twin trans- fusion syndrome (TTTS). Although fetoscopy increases survival rate, it O27 is also considered invasive and responsible for fetal and maternal Reasons of different therapeutic efficacy of HIFU ablation for complications that can affect neonatal outcome. These complications uterine fibroids with different MRI-T2W signal: thermal, acoustic are partially associated with the opening of the amniotic cavity. A and histopathological properties study completely non-invasive treatment that could occlude deep anasto- Faqi Li, Haoran Huang, Jianbo Ran, Zonggui Chen, Man Luo, Fei Li, moses would prevent the risks of invasive fetoscopy while offering Qi Wang, Jie Xu, Huan Liu, Zhibiao Wang the potential for more effective therapy. The efficacy of high-intensity College of Biomedical Engineering, State Key Laboratory of Ultrasound focused ultrasound (HIFU) is clinically proven for non-invasive tissue Engineering in Medicine Co-founded by Chongqing and the Ministry of ablation. We previously developed a toroidal HIFU transducer that Science and Technology, Chongqing Key Laboratory of Ultrasound in enables the destruction of large tissue volumes. In a recent study, we Medicine and Engineering, Chongqing Medical University, Chongqing, demonstrated the ability to induce lesions in human placenta using China this toroidal-shaped transducer without damaging intervening tissues Correspondence: Faqi Li within an ex vivo model. The effectiveness of this HIFU device ap- Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O27 plied to the perfused placental unit must be studied in a preclinical animal study under conditions similar to those in humans before OBJECTIVES To explore the causes of the differences in the thera- starting a clinical trial. Here, we report a feasibility study using a peutic efficacy of HIFU ablation for uterine fibroids with different T2- monkey model of pregnancy. The 3 objectives of this work were (i) weighted MRI signals from the perspective of thermal, acoustic and to evaluate the feasibility and reproducibility of HIFU lesions in the histopathologic characteristics. placenta of pregnant monkeys in vivo; (ii) to study the lesions using METHODS Totally 47 uterine fibroid specimens were collected after sur- ultrasound images, gross pathology, and microscopy; and (iii) to gery. According to the preoperative T2-weighted MRI signal, the uterine evaluate local andgeneral tolerance to the treatment. fibroids were classified into four categories, which were hypo-intense, iso- METHODS A toroidal HIFU transducer working at 2.5 MHz and com- intense, heterogeneous hyper-intense and homogeneous hyper-intense. posed of 32 ring-shaped emitters was used. An ultrasound probe work- The T2-weighted MRI signal of preoperative uterine fibroid was analyzed ing at 7.5 MHz was placed in the center of the HIFU transducer. The quantitatively to the signal intensity. Part of the uterine fibroids speci- imaging plane was aligned with the HIFU acoustic axis. The acoustic pa- mens were undergone the sound velocity, sound attenuation, specific rameters used during HIFU exposures were selected according to pre- heat capacity and thermal conductivity measurement, and others were liminary simulations taking into account the attenuation coefficient of used to analyze the content of tissue smooth muscle cell (SMC) and col- placentas, measured previously. In vivo experiments were then per- lagenous fiber (CF) by histopathological observation. HIFU irradiation was formed. Eight pregnant monkeys were exposed to HIFU treatment after made in the uterine fibroids specimens. Besides, with the clinical applica- anesthesia. Lesions on placental tissues were performed non-invasively tion and follow-up survey, the therapeutic effect of HIFU treatment for by placing the HIFU probe on the skin. Fetal and maternal parameters, the uterine fibroids was analyzed retrospectively. such as maternal heart rate, fetal heart rate, and subcutaneous and RESULTS There were the differences among the sound velocity, intra-amniotic fluid temperature, were recorded during HIFU exposure. sound attenuation, specific heat capacity and thermal conductivity of A C-section was performed immediately after insonification to extract the four groups of uterine fibroids, as well as the content of tissue the placenta, inspect the fetus and inspect the maternal abdominal smooth muscle cell and collagen fiber. The therapeutic efficacy has cavity. Placental HIFU lesions were studied using ultrasound images, significant differences in four groups of uterine fibroids, the groups gross pathology and histology. from the hypo-intense to homogeneous hyper-intense, the difficulty RESULTS The average gestational age of the animals was 72±4 days. of HIFU ablation have been gradually increased. There were correla- Thirteen HIFU exposures were performed on placentas. The parame- tions between the signal intensity, acoustic, thermal properties, histo- ters used in this study wereacoustic powers of 65, 80, 110 and 120 W pathological features and therapeutic efficacy of HIFU ablation for applied for 30, 15, 20 and 20 seconds, respectively. This gradual in- the uterine fibroids of different T2-weighted MRI signal. crease in the total energy delivered was used to determine aset of CONCLUSIONS There are some differences in the sound velocity, parameters to create reproducible lesions in the placentas without sound attenuation, number of SMC and CF content among the uter- any complications. Six placental lesions were observed with average ine fibroids of different MRIT2WI signal, which may be one of the im- diameters of 6.4 ± 0.5 mmand 7.8 ± 0.7 mm and an average depth portant factors to different therapeutic efficacy of HIFU ablation. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 18 of 122 (This study was supported by the National Natural ScienceFounda- avoid overexposure due to the heterogeneity of the tissues in the tar- tion of China (Grant Nos. 81127901, 11574039, 11274404)) get volume with significant variation in local absorption. open-loop control results, not shown, often resulted in either overexposure or underexposure depending on the value of the initial HIFU intensity. O28 The results also demonstrated the safety of the procedure as all the ani- Image-guided ablation of the carotid body in vivo: a potential mals tolerated this procedure well (over 32 animals at thetime of writ- noninvasive treatment for hypertension ing). Overall, the results show the feasibility of delivering prescribed 2,1 1 1 Dalong Liu , Raj Aravalli , Emad S. Ebbini levels of HIFU exposure to extremely small neurovascular structures Electrical and Computer Engineering, University of Minnesota, with potential application in the treatment of hypertension, targeted Minneapolis, Minnesota, USA; Siemens, Seattle, Washington, USA neuromodulation and other image-guided interventions. Based on the Correspondence: Dalong Liu results reported herein, a proof-of-concept study in 56 normotensive Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O28 and spontaneously hypertensive rats is currently under way. OBJECTIVES The carotid body (CB) is a neurovascular structure lo- cated near the carotid bifurcation. It is an arterial chemoreceptor, which produces neural output reflecting the partial pressure of O2 and CO2 as well as pH and temperature. The CB chemoreflex-evoked sympatho-excitatory responses are known to be enhanced in both human patients and animal model of hypertension. The main aim of this study is to establish the feasibility and safety of localized abla- tion of the carotid body usingimage-guided focused ultrasound (IgFUS) in normotensive and hypertensive rat models. METHODS Normotensive and hypertensive rats (275 - 330 gm) were treated using our dual-mode ultrasound array (DMUA) system for IgFUS using IACUC-approved protocol. In each experiment, the rat was placed in supine on stereotaxic stage with a heating pad placed in between to regu- late body temperature. The hair was shaved and removed using depilatory cream. The DMUA was positioned using a 3 stage motor under real-time imaging guidance. Synthetic-aperture (SA) Imaging was performed to lo- cate the carotid artery bifurcation based on a 3D scan of the treatment volume containing the target CB. The vessel pulsation was used to track the common (CCA), external (ECA) and internal (ICA) arteries in each plane. Once the carotid bifurcation plane was determined, four (4) treatment planes were definedoffset by 0.5, 1, 1.5 and 2 mm from the bifurcation in the caudal-to-rostral direction. In each plane, IgFUS was delivered at 2 - 6 Fig. 1 (abstract O28). See text for description locations around the ECA using a closed loop dose control (CLC) algorithm based on real-time DMUA imaging feedback (described previously at ISTU2015). Briefly, the echogenicity changes indicative of cavitation and/or tissue boiling are detected using DMUA imaging at up to 500 frames per second. These changes are used to adjust the HIFU exposure from frame to frame to avoid over-exposure to the target. The locations of the HIFU shots were chosen to assess the nature and extent of damage due to a "prescribed HIFU dose" under CLC in and around the target volume, in- cluding muscle and connective tissues. For each shot, the initial in situ HIFU intensity was between 5 - 10 kW/cm2, but this was quickly adjusted down by the CLC upon detection of the echogenicity change in beam- formed DMUA imaging data. The exposure time was 0.5 sec for each shot with minimum 40 ms at the initial intensity to ensure the echogenicity change is due to HIFU-induced change. Animals were survived for 48 - 96 hours after the treatment to allow for histological evaluation of HIFU-in- duced lesions (H&E) and the affected neural structures (Chromagranin A). Histology sections corresponding to the DMUA imaging slices during treat- ment were produced (50 μm per section covering the treatment volume.) RESULTS DMUA imaging data was collected before, during and after each HIFU shot at frame rates between 200 - 500 fps to document the lesion detection capabilitiesof the real-time guidance system. The system also logged the instantaneous intensity values used by the Fig. 2 (abstract O28). See text for description CLC algorithm. Figure 1 shows an example guidance and monitoring display from our DMUA system. The false color overlay represents the echogenicity change due to a HIFU shot placed just to the left of the right ECA nearthe bifurcation. The spatiotemporal maps (tem- poral-axial and temporal-lateral) demonstrate the high degree of O29 localization of the echogenicity change in DMUA imaging. The line Computer-aided transcranial ultrasound for time-efficient blood- profiles show the echogenicity change (in dB) at three points corre- brain barrier opening in primates 1 1 1 1 , Christian Aurup ,Carlos J. SierraSánchez , Julien Grondin , sponding to the target (red), vessel wall (blue) and skin (orange). Fig- Shih-Ying Wu 1 2 1,3 Wenlan Zheng , Vincent Ferrera , Elisa E. Konofagou ure 2 shows a typical H&E histology slide showing several discrete HIFU-induced shots (arrow heads) around the ECA. Biomedical Engineering, Columbia University, New York, New York, USA; Neuroscience, Columbia University, New York, New York, USA, CONCLUSIONS The results confirmed our ability to precisely control Radiology, Columbia University, New York, New York, USA the characteristics of HIFU-induced lesions very close to the vessel walls Correspondence: Shih-Ying Wu without damaging the endothelium, i.e. no danger of vessel perfor- ation. Furthermore, the results confirmed that CLC was essential to Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O29 Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 19 of 122 OBJECTIVES Focused ultrasound (FUS) with microbubbles holds great OBJECTIVES One third of the worldwide disease burden is caused by promise to noninvasively treat central-nervous-system diseases such as brain diseases, such as dementia, Parkinson’s and brain cancer. Des- Parkinson’sand Alzheimer’s disease through blood-brain barrier (BBB) pite efforts to develop new treatments, there remains no cure for opening and enhanced delivery of pharmaceuticals. While research has these diseases. A major reason for the lack of therapies is that most shown the need of repetitive application for treatment efficacy, currently drugs cannot enter the brain, because they are blocked by the there exists no clinical system designed to satisfy the requirement. In fact, blood-brain barrier (BBB). A promising solution uses focused ultra- the commonly-used magnetic resonance-guided FUS (MRgFUS) system sound and microbubbles to noninvasively and locally increase the hinders the re-application owing to its low speed, flexibility, and high cost BBB permeability so that drugs can enter the brain. This technology due to the dependence upon the MRI scanner. Therefore, in this study, a has successfully delivered a wide range of drugs, such as antibodies time-efficient transcranial FUS and monitoring system has been devel- and nanoparticles, and is currently being tested in human patients oped using computer-aided neuronavigation without requiring an online for the treatment of brain cancers. Despite encouraging results, the MRI, with the protocol from in silico planning, real-time targeting and current technology generates a poor drug distribution and too much monitoring, to post-treatment assessment evaluated in non-human pri- toxicity and damage for use in other brain diseases, such as demen- mates in vivo in preparation for clinical trials. tia, which affects 46 million people worldwide. Our group has re- METHODS An arm-free, real-time neuronavigation system designed for cently shown in vitro that these limitations may be due to the primates (both monkeys and humans) was customized to be used for conventional ultrasound sequences used to disrupt the BBB. These FUS application and monitoring purposes. The ultrasound system con- sequences consist of long-pulses emitted at a slow rate and generate sisted of a FUS treatment unit controlled by a customized program in a mixture of both desired and undesired cavitation activities. We Matlab with a single-element, 0.5-MHz FUS transducer triggered by a have recently developed and tested a new low pressure rapid short- function generator after 50-dB amplification, and an acoustic monitor- pulse (RaSP) sequence in vitro, designed to promote the desired cavi- ing unit with a programmable acoustic signal acquisition system and tation activity in the correct locations (e.g., capillaries). This new se- an array of acoustic detectors synchronized with the FUS system for quence is evaluated here for its ability to improve the in vivo real-time passive acoustic mapping and storage of the entire acoustic efficiency and safety of ultrasound-mediated drug delivery to the signals. Both the FUS and acoustic mapping were guided with the neu- brain. ronavigation system during the FUS procedure. The system was tested METHODS Rapid short-pulse (RaSP) sequences consist of short in 4 rhesus macaques with BBB opening in both sedate (animal under pulses separated by off-time intervals in the range of μs. In vitro anesthesia lying prone on an operating table) and awake setups (ani- studies have shown that these sequences prolong the lifetime of mal trained to sit in a customized chair), and the demonstrated proto- microbubbles and increase their mobility during the off-time in- col covered from in silico preplanning and simulation, real-time tervals, allowing lightly stimulated microbubbles to freely distrib- targeting and acoustic mapping guided by neuronavigation, to post- ute throughout the capillaries, which are the target treatment assessment of BBB opening effectivenss and safety in the microstructures of drug transfer. The better spatial and temporal MRI including contrast-enhanced T1-weighted imaging, T2-weighted distribution of microbubble activity allows a more uniform en- imaging, and susceptibility-weighted imaging (SWI). hancement of the BBB permeability and therefore a more uni- RESULTS Simulation revealed inter-animal variation (21%) due to varying form distribution of the delivered drug. We tested here whether skull density and thickness and can be compensated before sonication to RaSP sequences used in vivo in C57BL/6 mice would improve the ensure treatment reproducibility. In the in vivo experiments, for the first efficiency and safety of brain drug delivery. We compared our time the noninvasive FUS treatment was achieved within 30 min outside RaSP sequence (peak-negative pressure: 400 kPa; pulse length the MRI system for primates, and targeting flexibility allowed BBB open- (PL): 5 cycles; pulse repetition frequency (PRF): 1.25 kHz; burst ing in both the peripheral cerebral cortex and deeply-seated subcortical length: 10 ms) to the current gold standard, conventional se- structures with the use of a free-guide arm and the inflatable bladder sys- quenceat thesameacousticpressure(PL: 10,000cycles; PRF: 0.5 tem of the FUS transducer to couple with the scalp. Moreover, the Hz; burst length: 10 ms). Fluorescently-tagged (Texas Red) 3 kDa achieved targeting accuracy were proximal to the predicted 2-mm limit dextran and microbubbles were intravenously injected in mice in simulation. The accuracy in the awake animal setting (3.0 mm) was while sonicating the left hippocampus with a 1 MHz focused found to be comparable to the sedate animal setting (3.2 mm), which ultrasound transducer. A 7.5 MHz passive cavitation detector cap- was higher compared to the frame-based stereotaxis due to the improve- tured the microbubble acoustic emissions. The relative dose and ment of lateral positioning of the animal, and the focal shift in the acous- distribution of the drug were quantified by calculating the nor- tic beam path was due to the skull distortion. On the other hand, real- malised optical density (NOD, average increase in fluorescence in time cavitation mapping was performed with neuronavigation guidance the targeted area normalised by the control) and the coefficient during the entire sonication, with the frequency spectra data showed a of variation (COV, standard deviation over the average fluores- dramatic increase of the cavitation signal (harmonics and ultraharmonics) cence intensity in the targeted region). Safety was assessed by after injecting microbubbles. The acoustic mapping provided real-time haematoxylin and eosin (H&E) histological staining. spatial monitoring during the sonication and successfully confirmed the RESULTS Despite emitting 150 times less acoustic energy, RaSP location of BBB opening. Finally, in all the experiments performed, no sequences delivered a dextran dose of the same order of mag- acute damage such as hemorrhage (SWI) or edema (T2-weighted im- nitude as the long-pulse sequences (Fig. 2). Moreover, the drug aging) was detected upon radiologic examination 2 h after sonication. distribution was significantly more uniform using RaSP se- CONCLUSIONS This computer-aided system developed enabled rapid quences (Fig. 1), as indicated by the coefficient of variation (Fig. and flexible transcranial FUS applications for primates outside of the 2). Compared to conventional sequences, RaSP sequences pro- MRI scanner without the use of a stereotactic frame. It will greatly fa- duced less arterial effects, such as dextran accumulation in or cilitate both preclinical and clinical use for BBB opening and drug de- alongthe arteries (Fig.1c).This suggests that RaSP sequences livery to treat neurodegenerative disease and brain tumors. may reduce the likelihood and magnitude of unnecessary treat- ment of the arteries. Acoustic emissions from our short-pulses were more stable than those from the longpulses, with the en- ergy smoothly decreasing over time. The energy levels of the O30 acoustic emissions from the exposure to long-pulses varied Rapid short-pulse (Rasp) sequences – A new therapeutic greatly between pulses and within each pulse. ultrasound exposure paradigm to enhance drug delivery to the brain in vivo CONCLUSIONS These results suggest that RaSP sequences can im- Sophie Morse, Antonios Pouliopoulos, Tiffany Chan, Julien Lin, prove the spatial distribution of dextran delivery to the brain by pro- James J. Choi ducing a more uniform distribution of acoustic cavitation activity. Bioengineering, Imperial College London, London, UK Low pressure RaSP sequences could deliver a more efficient and safe Correspondence: Sophie Morse dose of drugs across the blood-brain barrier to treat diseases such as Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O30 Alzheimer’s, Parkinson’s and other neurodegenerative diseases. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 20 of 122 This abstract is not included as it has already been published: Nappoli A, Zaccagna F, Catocci G, Giulia B, Caliolo G, Andrani F, Catalano C. Magnetic resonance guided focused ultrasound surgery (MRgFUS) treatment of osteoid osteoma: a prospective development study. J Ther Ultrasound. 2015; 3(Suppl 1): O44. Available from: https://jtultrasound.bio- medcentral.com/articles/10.1186/2050-5736-3-S1-O44 O32 Transoesophageal HIFU for cardiac ablation: experiments on beating hearts 1 2 1 2 Paul Greillier , Bénédicte Ankou , Ali Zorgani , Francis Bessière , 3 3 4 4 Fabrice Marquet , Julie Magat , Sandrine Melot-Dusseau , Romain Lacoste , 3 5 2 1, 6 Bruno Quesson , Mathieu Pernot ,PhilippeChevalier , Cyril Lafon 1 2 LabTau - U1032, INSERM, LYON, Rhône, France; Hôpital Louis-Pradel, Lyon, 3 4 France; IHU-LIRYC - CHU Bordeaux, Pessac, France; Station de primatologie -CNRS- UPS846, Rousset, France; Institut Langevin - Ondes et Images - ESPCI ParisTech, CNRS UMR 7587, Paris, France; University of Virginia, Charlottesville, Virginia, USA Correspondence: Paul Greillier Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O32 Fig. 1 (abstract O30). In vivo brain drug delivery distributions using This abstract is not included as it has already been published: (a, b) rapid shortpulse (RaSP) or (c, d) conventional sequences. The Greillier P, Ankou B, Bessière, Zorgani A, Pioche M, Kwiecinski W, (a, c) left hippocampus of the mouse brain was sonicated in vivo Magat J, Melot-Dusseau S, Lacoste R, Quesson B, Pernot M, Catheline after systemic delivery of fluorescentlylabelled 3kDa dextran and S, Chevalier P, Lafon C. A75 Trans esophageal HIFU for cardiac abla- microbubbles. The (b, d) right hippocampus was not treated and tion: first experiment in non-human primate. J Ther Ultrasound. 2016; acted as our control (b, d). Sonication using (a) a RaSP sequence 4(Suppl 1):31. Available from: https://jtultrasound.biomedcentral.com/ delivered a greater dose of dextran to the brain parenchyma with articles/10.1186/s40349-016-0076-5 less delivery to arteries than (c) conventional sequences O33 In-vivo investigation of the combination of focused ultrasound and radiotherapy, using photoacoustic imaging as aplanning and monitoring tool Marcia M. Costa, Anant Shah, Ian Rivens, Tuathan O'Shea, Carol Box, Jeff Bamber, Gail ter Haar Radiotherapy and Imaging, The Institute of Cancer Research, Sutton, UK Correspondence: Marcia M. Costa Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O33 OBJECTIVES Tumour hypoxia is a limiting microenvironmental factor for radiotherapy (RT) success, since decreased oxygenation renders cells radioresistant, whereashigh intensity focused ultrasound (HIFU) tissue ablation is independent of oxygen levels. Imaging blood oxy- genation, which in part measures hypoxia, can be achieved non-inva- sively using photoacoustic imaging (PAI). The technique relies on the generationof acoustic waves (1-50MHz) by the tissue upon absorp- tion of short pulses of light. These waves can be detected using an ultrasound transducer and an image of relativeoptical absorption, Fig. 2 (abstract O30). Quantification of the normalised drug dose can then be reconstructed providing anatomical information. Further- and drug distribution delivered using RaSP and conventional more, light is absorbed differently by oxy- (HbO2) and deoxy-haemo- sequences. The normalised drug dose was quantified with the globin (Hb), and it is thus possible to calculate their relative normalised optical density (NOD), i.e. the average increase in proportions and, consequently, the distribution of oxygen saturation fluorescence in the targeted area normalised by the control. The (sO2=HbO2/(Hb+HbO2)) in tissue [1]. In this paper, we describe a drug distribution was quantified with the coefficient of variation combinatorial approach of HIFU and radiotherapy, using the former (COV), i.e. the standard deviation over the average fluorescence to target hypoxic (radioresistant) tumour volumes, in order to investi- intensity in the targeted region (* p < 0.01; ns = not significant) gateif this improves treatment outcome. PAI was used for HIFU treat- ment planning and assessing the treatment ouctcome. Furthermore, the relationship between the pretreatment sO2 values of tumours and radiotherapy response was also investigated. METHODS This study used a subcutaneously implanted human head O31 and neck tumour (CALR) in immunosuppressed mice, which was Magnetic resonance guided focused ultrasound surgery (MRgFUS) found to produce relativelyhypoxic tumours at a tumour size suitable Treatment of osteoid osteoma: a prospective development study for studies. Ultrasound imaging guided HIFU-alone (N=12) and com- Maayan Kimhy, Alessandro Napoli, Marco Marra Marcozzi puted tomography imaging guided radiotherapy-alone (N=13) treat- Dipartimento di Scienze Radiologiche, Oncologiche E Anatomo- ments were performed using dedicated small animal platforms: VIFU- Patologiche, Sapienza Università di Roma, Rome, Italy 2000 (Alpinion, Washington, USA) and SARRP (225KeV manufacturer Correspondence: Maayan Kimhy details, Xstrahl, Camberley, UK), respectively. Tumours had a mean Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O31 volume of 209 (+/-23 mm3). PAI (Multispectral Optoacoustic Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 21 of 122 Tomography (MSOT), iThera, Munich) was used to measure blood dis- tribution and its oxygen saturation non-invasively with the aim of identifying a main hypoxic tumour region for HIFU exposure. The HIFU treatment regime was a ‘spiral in’ pattern of 6 exposures, ISPTA -2 (free field) = 1.1 kW.cm and 10 second duration, per exposure. Exposures were done 1mm apart, in two rows, 3 exposures per row. Uniform whole tumour radiation exposures of 10 Gy, in a single fraction, were used. Animals were followed-up 60 days after treatment, using calliper measurements to calculate tumour volume. Once the treatment responses for individual treatments had been established, HIFU and RT were combined (total N=25), these being delivered 15 minutes apart. Both RTfollowed by HIFU and (N=12) the reverse treatment order (N=13) were investigated. RESULTS The HIFU-alone treatments resulted in 6 animals with no Fig. 2 (abstract O33). Treatment response A: proportion of tumour control and another 6 with an average tumour growth delay responders (R) and non-responders (NR) of 10gy-alone and combined of 6 (+/-4) days. The resultant hypointense lesion appearance in the treatments, showing an increase of responders for the latter group. B: Photoacoustic Imaging scan immediately after treatment is shown in Survival curves show an improvement in the overall survival of animals Fig. 1B. Surrounding the lesion, is a hyperintense region which spec- treated with the combined treatment compared to RT treatment alone. tral analysis showed to have increased haemoglobin content com- Arrows point to the average survival time of each treatment group pared to the pre-HIFU imaging (Fig. 1A). It is possible that this increase in haemoglobin signal is due to resultant haemorrhage (vas- cular disruption) caused by HIFU. Further histological studies are be- O34 ing undertaken to confirm these results. Radiotherapy alone Intracellular vaporization of antibody - conjugated phase - change treatments resulted in a 46% treatment response, defined as full nano-droplets for selective cancer therapy 1 1 2 1 tumour remission or growth inhibition within the follow up-period. Ayumu Ishijima , Takashi Azuma , Kosuke Minamihata , Satoshi Yamaguchi , 1 1 1 1 Spectral analysis of the photoacoustic signal showed a statistically Shinya Yamahira , Etsuko Kobayashi , Mariko Iijima , Yoshikazu Shibasaki , 1 1 significant difference between the sO2 values of responders (high TeruyukiNagamune ,IchiroSakuma 1 2 sO2) and non-responders (low sO2). The combined treatments out- The University of Tokyo, Tokyo, Japan; Kyushu University, Fukuoka, come, compared to the 10Gy-alone, showed an improvement in Japan treatment response both for percentage of responders (32% for RT Correspondence: Ayumu Ishijima +HFU and40% for HIFU+RT) and survival (Fig. 2). Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O34 CONCLUSIONS This preliminary study in a cell line known to be rela- tively radioresistant has shown a therapeutic benefit from combining OBJECTIVES Recent advances in nanomedicine provide the oppor- hypoxia targeted HIFU with whole tumour, in terms of improved out- tunity to reduce systemic toxicities of chemotherapies. However, come at lower doses. This effect is possibly due to the combination drug resistance remains a major challenge in cancer treatment re- of ablating radioresistant hypoxic areas and using RT for further kill search. Here we developed a nanomedicine composed of a phase- of the cells in better vascularised margins. Furthermore, PAI analysis change nano-droplet (PCND)1,2 and an anti-cancer antibody (9E5), showed the possibility of using this technique in adapting the treat- proposing the concept of ultrasound cancer therapy with intracellular ment planning forradiotherapy in clinic, as a relationship between vaporization to physically treat cancer. the sO2 measured values in tumour and the success of treatment METHODS 9E5-conjugated PCNDs (140 ± 120 nm) consists of a PFC was established. This project could have a major positive impact on liquid core (a mixture of perfluoropentane and perfluorohexane), a the treatment of tumours that are characteristically hypoxic, such as phospholipid shell, and antibody 9E5. The 9E5 human anti-epiregulin those of head and neck, if such combinationtreatments can be trans- (EREG) antibody was selected for active targeting of PCNDs. EREG, lated into clinical practice. [1] Xia et al., Electromagn Waves (Camb). the cell-membrane–expressed ligand of epidermal growth factor re- 2014; 147: 1–22. ceptor, is expressed and integrated into the plasma membrane at relatively high levels in a variety of human cancers.3 9E5 was conju- gated to PCNDs using the biotin streptavidin-biotin binding tech- nique. Biotinylated-9E5 and Alexa Fluor 647-conjugated streptavidin (SA-AF647) were bound to biotinylated PCND.The human colonic adenocarcinoma cell line DLD1 and the human gastric cancer cell line AGS were selected as high and low EREG-expressing cancer cell lines, respectively. To demonstrate the selective targeting capability of 9E5-conjugated PCNDs to DLD1 cells, we used SA-AF647 as a fluorescence probe. The targeted cells were observed by confocal laser scanning microscopy (CLSM), and the number of PCNDs at- tached and the fraction of bound DLD1 cells were quantitatively measuredby flow cytometer. Next, PCND-accumulated cells were ex- posed to ultrasound (100 cycles, 4 MHz, a peak negative pressure of 1.5 MPa), and its cytotoxic effects were visualized. The intracellular vaporization of 9E5-conjugated PCNDs in DLD1 cells was observed by a high-speed imaging system recording 101 subsequent frames at 0.25 Mfps. Furthermore, we quantitatively evaluated the cytotoxic Fig. 1 (abstract O33). A: Pre-HIFU photoacoustic imaging scan of R capabilities of vaporized PCNDs using PI staining and flow cytometer CAL tumour. B: Post-HIFU scan showing the lesion (red arrow) and after exposure of 5 cycles at 5 MHz of pulsed ultrasound with a peak a hypertensity rim (blue arrow) negative pressure of 4.6 MPa using an ultrasound imaging probe. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 22 of 122 RESULTS CLSM images of DLD1 cells treated with 9E5-conjugated OBJECTIVES The utility of Time Reversal Cavities (TRC) for medical PCNDs show the fluorescence signal inside the cells after 3 hours of ultrasound has been demonstrated for several applications. They incubation (Fig. 1a), whereas no clear fluorescence signals were ob- have indeed shown promising results for shock wave therapy, as served from the other types of cells. Flow cytometry analysis (N = 5) they allow high intensity pulses focusing in large regions of interest indicated that the 97.8 ± 0.5% of DLD1 cells were targeted by 9E5- with only electronical steering and a limited number of transducer el- conjugated PCNDs, whereas 1.4 ± 0.3% of DLD1 cells were targeted ements [1, 2]. TRCs have also been used as compact emitter- by non-9E5-conjugated PCNDs, similar to that of the control (4.4 ± receivers for pulse echo imaging [3 - 5]. In this work we present a 1.6%). Furthermore, the ratio ofAGS cells targeted by 9E5-conjugated dual mode cavity that can both perform shock wave therapy and 3D and non–9E5-conjugated PCNDs were close to the levels of the con- imaging in a large region of interest with 128 transducer elements. trol. The intracellular vaporization observed by high-speed imaging This dual mode capability is particularly interesting in transcostal ap- revealed that cell membranes were ruptured or broken into several plications where part of the therapeutic beam is shadowed by the parts during this initial stage ofvaporization (Fig. 1b). High-speed im- ribs lowering significantly the focal pressure. 3D volumetric imaging ages clearly show that intracellular vaporization caused a significant of the ribcage is used to perform an adaptive focusing of the thera- disturbance in cell morphology and destroyed the cells. The fraction peutic beam by transmitting selectively the ultrasonic energy of viable cells after ultrasound exposure was measured by flow cy- through the ribs. Adaptive focusing results in an increased efficiency tometry (N = 5). Cell viability was significantly reduced to 43.0 ± and higher pressure at the focal zone. 5.6% for 9E5-conjugatedPCNDs-treated DLD1 cells, while there was METHODS The leaky TRC is made of 2 orthogonal steel rod forests in no significant cell viability decrease for PCNDs without 9E5 conjuga- a reverberating cavity. The cavity itself is 20*5*5 cm with steel walls, tion (95.8 ± 2.0%) and without ultrasound exposure. Furthermore, and filled with water. A high power matrix array transducer (128- the viability of AGS cells did not decrease (98.0 ± 0.2%). These data elements, 1 MHz, Imasonics, France) is placed in the back of the indicate that PCND conjugated with 9E5 can sufficiently kill DLD1 cavity, opposite the aperture. For shock wave therapy experiments, cells with high selectivity. The addition of free 9E5 to DLD1 cells be- the probe was driven by custom multi-channel electronics (Corelec, fore treating/co-treating with 9E5-conjugated PCNDs significantly re- France). 40 μs US pulses emitted through the cavity were spread to duced the number of PI-stained cells (89.5 ±10.2%). up to 1 ms, picked up by a HNC 200 hydrophone (Onda, Sunnyvale, CONCLUSION In order to obtain cytotoxicity with droplet CA) and stored. This process will further on be called calibration step. vaporization, previous reports combined anti-cancer drugs such as Time reversal focusing (TRF) by compressing these signals in space doxorubicin with droplets.4 Our studies revealed that anticancer and time allowed us to reach the needed pressures. Steering the drug free 9E5-conjugated PCNDs are selectively internalized into tar- focal spot over a large volume was achieved by moving the hydro- geted cancer cells and kill the cells dynamically by ultrasound-in- phone (42*42 mm2 area, 1.5 mm grid step). We then re-emitted the duced intracellular vaporization. In vitro experiments show that 9E5- reversed signal at a pulse repetition frequency of 260 Hz, targeting conjugated PCND targets 97.8% of high EREG-expressing cancer cells an Ultracal phantom to form lesions. For imaging experiments, the and kills 57% of those targeted upon exposure to ultrasound. Fur- probe was driven by a Verasonic HIFU system. The same calibration thermore, direct observation of the intracellular vaporization process step was used, but instead of TRF, inverse filter (IF) focusing was revealed the significant morphological alterations of cells and the re- used. The focus quality was assessed by hydrophone scanning. The lease of intracellular contents. This work was supported by a grant signals corresponding to a focus on each grid point were succes- for the TSBMI from the MEXT of Japan. A.I. was supported by a sively emitted in free field, and the backscattered signals were re- Grant-in-Aid from the JSPS Research Fellows. corded on the transducers and stored, to constitute a reference library. An object was then placed in the focal plane, and the same References process was repeated. For each grid point, the reference was sub- 1. Kawabata K et al. Jpn J Appl Phys 2005; 44: 4548. tracted from the backscattered signals, and an image of the object 2. Ishijima A et al. Ultrasonics 2016; 69: 97–105. was reconstructed using the IF signals for focusing on reception. We 3. Lee YH et al. Biochem Biophys Res Commun 2013; 441: 1011–1017. used 2D simulations (Acel) to evaluate the pressure gain with select- 4. Wang CH et al. Biomaterials 2012; 33: 1939–1947. ive propagation of the therapeutic beam through the intercostal space. Calibration step on grid points in front of the cavity aperture was performed in free space. Reflectors mimicking a rib cage were then added in the propagating medium, blocking part of the aper- ture. The grid points were used as virtual transducers, and a delay law for a focus behind the ribs was applied. Either all the grid points or only those in the intercostal spaces were used. The total power was the same. Peak pressure was recorded on target in both cases. RESULTS Hydrophone measurements showed the device could cre- ate thin isotropic focal spots in a 200 cm region of interest, with very limited decrease of the peak pressure Therapy: Targeting the Ultracal® phantom in 7 different positions, we formed clear isotropic lesions in a 2.5x4 cm region, with only electronical steering. Imaging: The focal spots obtained with IF focusing showed central lobes with a 2 λ wide - 6 dB area, and a global -20 dB contrast. Side Fig. 1 (abstract O34). (a) Internalization of AF647labeled lobes were visible in one direction, but remained 10 dB below the 9E5conjugated PCND. CLSM observation of 9E5 antibodymediated main focus. We imaged either 2 steel wires, or 2 human ribs. In both accumulation and internalization of AF647labeled9E5conjugated cases, the object was clearly visible on the final image. The imaging PCNDs into DLD1 cells. (b) Highspeed imaging of intracellular vaporization window (4x4cm ) coincides with the device aperture. Coupling imaging and therapy: Simulations showed a 21 % increase in the focal peak pressure when turning off the virtual transducers placed O35 in front of the ribs Dual mode time reversal cavity for US shockwave therapy and 3D CONCLUSIONS We built a dual mode TRC that can both perform US imaging shock wave therapy and create a 3D image of the surrounding 1,2 1 1 1 J. Robin , B. Arnal , M. Tanter , M. Pernot medium. Though the image resolution is quite low due the low fre- 1 2 Institut Langevin, Paris, France; Université Paris 7, Paris, France quency used, it is sufficient to visualize strongly reflecting structures Correspondence: J. Robin like the rib cage. This could be particularly interesting in transcostal Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O35 therapy of the heart or the liver. Preliminary results indeed show that Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 23 of 122 selective propagation of the therapeutic beam through the intercos- O37 tal space would increase the peak pressure on target. Application of MR-ARFI to monitoring the stiffness changes of porcine muscle after HIFU therapy References Yangzi Qiao, Chao Zou, Xin Liu, Hairong Zheng [1] Arnal et al, Appl Phys Lett, 101 1-5, 2012 Shenzhen Institutes of Advanced Technology, Chinese Academy, [2] Robin et al, IEEE IUS, 2015 Shenzhen, China [3] Sarvazyan et al, Acoust Phys 55 630–7, 2009 Correspondence: Yangzi Qiao [4] Luong et al, Sci Rep 6 36096, 2016 Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O37 [5] Montaldo et al, Appl Phys Lett 2004 (No Image Selected) OBJECTIVES HIFU with the capability to treat deep tumor precisely is a fast developing therapeutic method. Due to the multiple contrast and unique thermometry ability, MRI has been the most popular im- O36 aging modality for HIFU guidance. Stiffness is an intrinsic property of The effects of steroids on the myocardial reduction induced by tissue, and can be monitored by MR-ARFI during HIFUtherapy. How- myocardial cavitation-enabled therapy (MCET) 1 2 2 2 2, 1 ever, the MR-ARFI based tissue stiffness evaluation results were quite Y. I. Zhu ,X.Lu , C. Dou , D. L. Miller , O. D. Kripfgans different in reports. It is expected that the tissue becomes stiffer after O.D. Kripfgans, Biomedical Engineering, University of Michigan, Ann coagulation, producing decreased displacement. But in some other Arbor, Michigan, USA; Department of Radiology, University of Michigan, reports, the coagulation or stiffer implant leads to an increased dis- Ann Arbor, Michigan, USA placement. The inconsistency of the reports may due to the different Correspondence: Y. I. Zhu acoustic parameters of different tissue. In this study, MR-ARFI was ap- Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O36 plied to monitoring the stiffness changes of porcine muscle during HIFU therapy. The results demonstrate that ARFI induced displace- OBJECTIVES Myocardial Cavitation Enabled Therapy (MCET) has been ment change is strongly correlated to coagulation. The displacement proposed as a means to achieve minimally invasive myocardial reduc- difference map can be used to depict the coagulation region, espe- tion using ultrasound to produce scattered microlesions by cavitating cially for small coagulation, making MR-ARFI an important comple- contrast agent microbubbles. Previous studies have demonstrated the mentary method for tissue monitoring during HIFU therapy. efficacy of MCET to induce damage. The purpose of this work was to METHODS All experiments were conducted on a 3T MR system (TIM study the treatment effect of the steroid methylprednisolone (MP), in Trio, Siemens). Ex vivo porcine muscle were sonicated by a 1.0MHz terms of swelling reduction following MCET and inhibition of fibrous tis- HIFU transducer (Imasonic). The input electric power was set from sue formation during long-term healing process. 8W to 100W, and transferred to the samples continuously in 10s to METHODS Dahl/SS rats from Charles River were used as an in vivo 60s. The segment SE-EPI (sEPI) sequence was usedfor ARFI imaging. model for HCM. Under ketamin/zylazine IP anesthesia, contrast agent The amplitude of displacement-encoding gradients (DEG) was 32mT/ was infused at a rate of 5μL/min/kg (tail vein catheter). The shaved and m, with input HIFU power of 55W and duty cycle of 2.5%. The dur- depilated left thorax was aimed at with a cardiac phased array (10S, ation of each DEG was 10ms. Vivid 7, GE Healthcare) to center on the left ventricular myocardium. In Imaging protocol was: TR=600ms, TE=36ms, slice thickness=5mm, this arrangement a 19 mm diameter single element therapy transducer resolution=1.6*1.6mm2, matrix=54*128, EPI factor=9. Acquisition was co-aligned to aim at a registered region of interest in the field of time=8.4s. Two sets of images with opposite polarity of DEG were view of the 10S array. For therapy 10-cycle tone bursts at 1.5 MHz, 4 used to quantify the displacement in each measurement. Ten mea- repetitions at 0.25 ms pulse interval, i.e. 4.0 kHz PRF, were sent every 8 surements of ARFI were scanned before and immediately after HIFU heartbeats, aligned with trigger end systole (RR/3, using ECG gating). sonication. T-test was used to determine whether tissue displace- Peak negative free field pressures of 4 MPa were used to induce cavita- ments have significantly changed. Temperature rise was monitored tion for 10 min. Therapy and sham therapy groups were followed up by GREduring HIFU sonication. The protocol was: TR=29ms, TE=10ms with MP administered at 0, 3, 6 and 24 hours after ultrasound exposure. with the same FOV and resolution. The ambient temperature was 19° Specifically, 30 mg/kg was chosen as high dose while 1 mg/kg was C. T2w image was acquired after HIFU sonication with TR=5000ms, used as low dose alternative. Myocardial wall thickness 24 hours after TE=89ms, resolution=0.8*0.8 mm2, matrix=108*256. therapy, measured from echocardiography was used to gauge the ef- RESULTS Figure 1(a) shows the displacement maps overlaid on ana- fect of initial myocardial swelling. White blood cell count was carried tomical image. No coagulation was found with 60 s sonication under out 24 hours after therapy. Hearts were removed after 4 weeks and ex- 32 W power. The peak temperature change was 17.3°C in the end of amined for evidence of the MP treatment effect. Histological sections the sonication, resulting in a peak temperature around 36°C. The av- with Masson’s trichrome staining were quantitatively analyzed for ex- eraged maximal displacement before sonication was 3.41±0.46 μm, tent of fibrosis, i.e. tissue scarring. while the averaged maximal displacement after sonication was 3.36 RESULTS Myocardial wall thickness from echocardiography was mea- ±0.26 μm. The maximal displacement was constant during the suc- sured as: sham 1.69±1.6 mm; therapy only 2.68±0.21 mm; therapy with cessive 10 measurements after sonication, indicating the maximal low MP 2.29±0.22 mm; and therapy with high MP 2.01±0.14 mm. High displacement might be independent of temperature change (Figure dose of MP had a 25% wall swelling reduction with a p-value of 0.003 1(b)). It is also shown in Figure 1(c) there is no signification difference relative to the therapy only group. Absolute neutrophil count demon- between the maximal displacements before and after HIFU sonic- strated the efficacy of MP: sham 5.0±1.1 109/L; therapy only 5.4±0.9 ation with p = 0.18 from t-test. Figure 2 shows another case with in- 109/L; therapy with low MP 3.6±1.4 109/L; and therapy with high MP put power 82 W under 30s sonication, coagulation occurred. The 2.7±1.2 109/L. High dose of MP had a 45% neutrophil count reduction peak temperature was around 85°C (peak temperature change was though with a p-value of 0.026 relative to the therapy only group. Fi- 66.7°C), much higher than the protein denature temperature. The dis- brosis densities were quantified for the treated regions as to represent placement maps before and after sonication differed a lot. The max- the extent of scarring with results therapy only 25.6±4.0%; therapy with imal displacement significantly increased after sonication, with an low MP 25.7±1.6%; and therapy with high MP 20.6±0.4%. High dose averaged increment of 1.86 μm (p<0.01). The averaged maximal dis- MP on average reduced fibrosis formation of 20% though is underpow- placement before sonication was 3.46±0.17 μm, while the averaged ered in this study with a p-value of 0.154. maximal displacement after sonication was 5.32±0.28 μm. T2w image CONCLUSIONS In this MCET rodent study, reduction trend of swelling and displacement difference map (the difference between the dis- wall thickness and neutrophil count has shown that MP is effective to placement before and after HIFU sonication) were compared for inhibit the inflammatory response and reduce swelling right after ther- visualization of thecoagulation. In T2w image, the coagulation was apy. Additionally, long-term study of fibrosis density analysis indicated pointed out by a red dashed circle (Fig. 3). In displacement difference that MP potentially helps reducing fibrosis formation. Cavitation en- map, the region where displacement before and after HIFUsonication hanced therapy maybe a possible avenue for non-invasive tissue reduc- show significant difference (p<0.01) was overlaid on the anatomic tion for treatment of hypertrophic cardiomyopathy. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 24 of 122 image. In another sonication (with input power 82W, 10s sonication), The peak temperature was around 48.2°C (peak temperature change was 29.2°C). However, the coagulation can be hardly recognized in T2w image, while it is still visible in displacement difference map. The averaged maximal displacement before sonication was 3.45±0.12 μm. While the averaged maximal displacement after sonication was 5.06±0.19 μm. The coagulation area is marked in Fig. 4(c). CONCLUSIONS In conclusion, MR-ARFI can detect significant changes of porcine muscle displacement while coagulation occurred. After heating the displacement significantly increased at the coagulation region. The displacement difference map can be used to visualize and evaluate therapy effect, especially for small coagulation. This makes MR-ARFI an important complementary method for tissue monitoring during HIFU therapy. Fig. 3 (abstract O37). (a) T2w image, (b) displacement difference map, and (c) photo of coagulation with input power of 82W, 30s continuous sonication Fig. 1 (abstract O37). (a) The displacement map before (left) and after (right) HIFU sonication. (b) The maximal displacement in focus. (c) The averaged maximal displacement with inputpower of 32W, 60s continuous Fig. 4 (abstract O37). (a) T2w image, (b) displacement difference map, and (c) photo of coagulation with input power of 82W, 10s continuous sonication O39 Long-term oncologic outcomes of focal magnetic resonance guided focused ultrasound treatment for locally confined prostate cancer Alexander Nosov, Sergey Reva, Maria Berkut Fig. 2 (abstract O37). (a) The displacement map before (left) and Onco-urological Department, N.N. Petrov Research Institute of Oncology, after (right) HIFU sonication. (b) maximal displacement in focus. (c) Saint-Petersburg, Russian Federation The averaged maximal displacement with input power of 82W, 30s Correspondence: Alexander Nosov continuous sonication Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O39 Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 25 of 122 OBJECTIVES Progress in different diagnostic and treatment mo- tremor score, and T2-weighted magnetic resonance imaging (T2- dalities of prostate cancer (PCa) have strengthened support for MRI). Each of these methods provides an imperfect estimate of the the use of focal high-intensity focused ultrasound (HIFU). How- ultimate treatment effect. Diffusion-weighted MR imaging (DW-MRI) ever, important questions remain regarding candidate selection, can potentially provide accurate lesion visualization and prediction at treatment, and outcomes. We assessed long-term oncologic out- earlier time points than T2-MRI. This is analogous toischemic stroke, comes of focalHIFU in a small single-center cohort of low-risk where DW-MRI is able to observe infarcts at early time points. To PCa patients. date, DW-MRI has achieved very limited success because of technical METHODS Twenty-two patients with low risk PCa (PSA<10 ng/ difficulties introduced by the focused ultrasound system such as ml, Gleason score less than 7, or clinical stage cT2b and less) eddy currents, magnetic field inhomogeneity, and patient motion. A were underwent for focal HIFU ablation (ExAblate 2100 for a dual-echo, steady-state sequence has found moderate success at prostate device, InSightec) with GE MRI suite and endorectal identifying lesions in a pig model [1]. Here, we propose a single-shot, FUS transducer from March 2009 to January 2010. Among them, spiral DW-MRI sequence coupled with an outer volume suppression 8 patients were available for long term (a median time – 7.3 preparation sequence [2] as a means to quickly obtain diffusion years) follow-up. Desired treatment target (Region of Treatment, weighted images of sub thalamic nuclei. See Fig. 1. The single-shot, ROT) and vulnerable structures (nerve vascular bundles) were spiral acquisition reduces sequence sensitivity to bulk motion and marked on acquired magnetic resonance guided (MRg) planning eddy currents. Meanwhile, the outer volume suppression pulse re- images. Pre- and post-reatment strategy, rate and follow-up duces the imaging field of view to a region that can be successfully schedule was designed at PCa001 andPCa002 studies and de- supported by the spiral trajectory. scribed previously; prospective parts of these studies was closed METHODS The proposed sequence is displayed in Fig. 1. A single- after 6-month follow up. shot, spiral acquisition is used during the readout portion of a Stejs- RESULTS The average patient’s age 64 (49-73) years. Median pre- kal-Tanner diffusion-weightedsequence. The spiral gradient samples HIFU PSA level and post-HIFU PSA nadir was 7.6 and 3.9 ng/ml, a 4 cm diameter circular plane at 1.5 mm resolution using a variable respectively. Biochemicalrecurrence (BCR, defined as nadir + 2 density trajectory of 10 ms duration. The sequence is prepended by ng/ml) was observed in 7 (87.5%) cases. Medium time to progres- an outer volume suppression pulse consisting of an 8 ms long BIR-4 sion was 18 (3-32) months. In 4 (50%) cases local progression adiabatic excitation pulse and a 4 ms long cylindrical tip-back pulse was confirmed by prostate biopsy and after metastatic process as described in [2]. The preparation pulse reduces the field of avail- exclusion salvage radical prostatectomy (RPE). Generally, surgery able magnetizaiton to a cylinder approximately 3.5 cm in diameter. after ablation was severe than in naïve patients; however, opera- This sequence was validated using the the bodycoil of a 3T MR scan- tive characteristics (operative time, blood loss, hospital stay) were ner (GE, Waukesha, WI) and a human volunteer. Imaging parameters comparable with historical cohort. During follow-up time, sys- are; TR: 1 s, TE: 45 ms, FOV: 5 cm, resolution: 1.5 mm, slice thickness: temic treatment (hormonal therapy) was prescribed for 5 patients, 0.8cm, bValue: 750 smm-2, averages: 16. We will measure the time as a result of distant metastatic progression after prostatectomy course of the diffusion-weighted signal after sonication by subjecting (2) and without secondary local treatment (3). Cancerspecific sur- two instances of a porcine model with craniotomy to focused ultra- vival (CSS) and overall survival (OS) was both 100%. soundablation (ExAblate Neuro, INSIGHTEC, Haifa, Israel) while under CONCLUSIONS Despiteacceptableresults of survival,wefound anesthesia. A total of 8 lesions will be made in each animal and DW that almost all patients are progressed during follow-up. These and T2-weighted images will be acquired at time points spaced by 2 data are in contrast with previously published data. However, minutes from immediately after sonication to 20 minutes after sonic- patients in our series were younger than in historical cohort. ation using the proposed sequence for DW-MRI and a fast-spin echo We concluded that the use of focal high-intensity focused ultra- sequence with spiral acquisitons for T2-MRI. We will then compare soundinselectedpatientsrepresents a strategy combining the time course of the DW-MRI signal immediately after ablation to benefit of active surveillance and radical treatment in patients the time course of the T2-MRI signal. with low risk PCa. However, this concept should be evaluated in RESULTS Figure 2A and B demonstrate the outer volume suppression large prospective controlled studies. sequence using an enlarged field of view and a six-shot, interleaved spiral acquisition. By applying the suppression preparation pulse, the field of view is successfully reduced to a region about the right ven- tricle. A single-shot spiral image with reduced field of view of this re- gion, as well as an identical image with diffusion weighting, are O40 displayed in Fig. 2C and D, respectively. Diffusion-weighted contrast Rapid diffusion-weighted imaging immediately after sonication can be observed in the suppressed signal of the ventricle. The poor using outer volume suppression pulses and single-shot, spiral transmit and recieve gains of the body coil do not prevent the suc- acquisition 1 1 1 2 cessful implementation of this sequence. Steven P. Allen , Xue Feng , Helen L. Sporkin , Jeff Elias , 3 1 CONCLUSIONS Single-shot, spiral acquisition with an outer volume Kim Butts-Pauly , Craig H. Meyer suppression preparation pulse shows promise as a motion-insensi- Biomedical Engineering, University of Virginia, Charlottesville, Virginia, tive sequence to capture diffusion weighted contrast in the USA; Neurological Surgery, University of Virginia Hospital, Charlottesville, thalamus. Virginia, USA; Radiology, Stanford University, Palo Alto, Virginia, USA Correspondence: Steven P. Allen Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O40 References [1] Plata et al. “A feasibility study on monitoring the evolution of apparent OBJECTIVES There remains a critical need to predict the long term diffusion coefficient decrease during thermal ablation,” Med. Phys., 42(8), effects of a sonication during MR-guided, focused ultrasound thala- 5130–5137, 2015 motomy. Current prediction methods include calculations based on [2] Smith et al. “Reduced field of view MRI with rapid, B1-robust outer the observed temperature trajectory, observations of the patient’s volume suppression,” Magn. Reson. Med., 67(5), 1316–1323, 2012. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 26 of 122 O41 Selection of MR-HIFU hyperthermia treatment sites based on MR thermometry evaluation in healthy volunteers 1 2 3 1 Satya V.V.N. Kothapalli , Michael Altman , Ari Partanen , Lifei Zhu , 1 2 2 2 Galen Cheng , H. Michael Gach , William Straube , Dennis Hallahan , 1,2 Hong Chen Biomedical Engieering, Washington University in Saint Louis, Saint Louis, Missouri, USA; Department of Radiation Oncology, Washington University in St. Louis, Saint Louis, Missouri, USA; Clinical Science MR Therapy, Philips, Andover, Massachusetts, USA Correspondence: Satya V.V.N. Kothapalli Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O41 OBJECTIVES Magnetic resonance imaging-guided high-intensity fo- cused ultrasound (MR-HIFU) is a promising noninvasive technique for mild hyperthermia, enabling targeted and localized hyperthermia Fig. 1 (abstract O40). The proposed diffusionweighted sequence. therapy deep within the body under real-time MRI-based An outer volume suppression pulse is prepended to a StejskalTanner temperature monitoring and control. However, the narrow thera- diffusionweighted imaging pulse using a singleshot spiral trajectory peutic window (41-43°C) and long treatment duration (30-60 mi- for image acquisiton nutes) pose a great challenge on MR thermometry. As a first step toward effective clinical translation of MRHIFU hyperthermia, this study identified preferable treatment sites based on the accuracy and precision of MR thermometry performed on healthy volunteers. METHODS A total of 15 healthy volunteers (age 18-45 y and body weight 45-90 kg) were recruited. A clinical MR-HIFU system (Sonalleve V2, Philips, Vantaa, Finland) was used with the patient tabletop docked into the MRI bore (Ingenia 1.5T, Philips, Best, the Netherlands). Volun- teers were positioned above the tabletop’sacousticwindow and a standard HIFU-compatible 3-ch pelvis RF receive coil was secured over the target anatomy. For MR image acquisition, the pelvis coil was used together with the 2-ch RF receive coil integrated in the HIFU tabletop. A dynamic multi-slice fast-field-echo pulse sequence (sequence (TR=41 ms; TE=19 ms;voxel=2.5×2.5×7 mm2; FOV=400×400 mm2)) was used for real-time MRI (without HIFU sonication) together with the proton reson- ance frequency shift (PRFS) method to calculate temperature maps. Eight volunteers were subjected to a shorter scanning period (5 mins) targeting the upper body, pelvis, and lower extremity. Seven volunteers were subjected to a longer scanning period (30 mins) targeting the lower extremities, i.e., thigh and calf. The precision of MR thermometry was quantified by the temporal temperature uncertainty, which calcu- lated the temporal standard deviation for each pixel within an 18×18 mm2 ROI. The accuracy of MR thermometry wasquantified within the same ROI by the absolute error of measured temperatures compared to body temperature (37°C). Temperature monitoring with both uncertainty and absolute error lower than 1°C is expected to be satisfactory for hyperthermia. RESULTS MR thermometry measurements based on 5-min scans at the chest, pelvis, and lower extremity had an uncertainty of 2.53°C ±0.48°C, 1.89°C±0.50°C, and0.50°C±0.04°C, respectively, and an abso- lute error of 0.63°C±0.63°C, 2.88°C±0.87°C, and 0.08°C±0.13°C, re- spectively. MR thermometry measurements based on 30-min scans Fig. 2 (abstract O40). Validation of the proposed sequence in at the lower extremity found the uncertainty and the absolute error a human volunteer using the body coil of a 3T MR scanner. of the MR thermometry to be 0.52°C±0.13°C and 0.12°C±0.06°C, re- (A) Image acquired with a nominal field of view. (B) The spectively. No significant difference was found between 5-min scan- same sequence as (A), but prepended with the outer volume ning and 30-min scanning for the lower extremity. Among the three suppression pulse. The preparation pulse reduces the region of anatomical sites, only the lower extremity had satisfactory excited magnetization to a small cylinder. (C) The reduced field temperature accuracy and precision according to the suggested of view imaged with a singleshot spiral acquisiton. (D) Same as criterion. (C) but with a diffusionweighting of 750 smm2. Diffusionweighting can CONCLUSIONS This study constitutes the first evaluation of MR be observed inthe suppression of the ventricle. The poor transmit and thermometry performance at different body sites for long scan recieve gains of the body coil do not prevent the successful times that are relevant in clinical MRgHIFU hyperthermia ther- implementation of this sequence apy. Motion induced by respiration and cardiac activity poses a Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 27 of 122 technical challenge on the application of this treatment on chest and pelvis. Theextremity was found to be a preferable site for MR-HIFU hyperthermia using the suggested criteria that both the uncertainty and absolute error need to be under 1°C. O42 Behaviour of bubbles generated by acoustic droplet vaporization within inter-tissue region Y. Ho, C. Yeh Department of Biomedical Engineering and Environmental Sciences, National Tsing Hua University Correspondence: Y. Ho Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O42 OBJECTIVES Treatment resistance is a critical consideration in cancer research. The tumor cells which extensively proliferate away from vessels show treatment resistance due to the restricted drug penetration and biological mutation of hypoxia. Acoustic droplet vaporization (ADV) can convert nanodroplets into gaseous bubbles (ADV-Bs) and simultaneously disrupt vessels to improves the penetration of nanodroplets, ADV-Bs, and drugs. Previous studies have been proved the mechanical Fig. 1 (abstract O42). See text for description force of bubble cavitation can directly induce in vitro and intra- vascular bioeffects. Therefore, our study investigated the behav- iors of ADV-Bs in vessels and tissue triggered by ultrasound O43 (US), and then evaluated the feasibility of using intertissue ADV- Enhanced ultrasound-mediated trans-membrane drug delivery into Bs to treat resistant tumor cells by physical damage. a monolayer of endothelial cells during high flow in vitro METHODS The window chamber model was used to directly observe conditions vessels and tissue pattern on mouse dorsal skin by the integrated Syed Mohd Nooh Syed Omar, Rob Krams, James J. Choi acousto-optical system. Mice were injected 30 μL NDs (304±97 nm) Bioengineering, Imperial College London, London, UK with the concentration of 1.2-1.4×10 NDs/mL to form ADV-Bs by 3- Correspondence: Syed Mohd Nooh Syed Omar cycle single-pulsed US at 10 MPa. The modulated parameters of US Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O43 with pulse repeat frequency of 1 Hz were used to stimulate ADV-Bs as following: (1) 100-cycle with 5 and 10 MPa for movement; (2) 3- OBJECTIVES One of the challenges in the treatment of atherosclerosis cycle with 5 MPa for cavitation. Furthermore, the transgenic adeno- and other cardiovascular disease is the inability to deliver therapeutic carcinoma of the mouse prostate (TRAMP) cells with live-cell imper- drugs across the cell membrane effectively and safely. Entire classes of meable dye propidium iodide (PI) were used to evaluate cellular drugs, nucleic acids (e.g., DNA, siRNA and mRNA) and biomolecules for im- bioeffect induced by ADV-Bs. proving treatment, are ineffective, because they cannot enter the cell at RESULTS Intravital images showed vascular disruption, intravas- safe systemic doses that do not cause side effects throughout the rest of cular ADV-Bs, and intertissue ADV-Bs after ADV. The intravascu- the body. Over the last several decades, ultrasound-mediated trans-mem- lar ADV-Bs were moved following the vascular shape and stuck brane drug delivery methods (e.g. sonoporation) have been studied at the branch (black arrows) during US stimulation (Fig. 1. A). in vitro to deliver drugs into cells. Through these studies, it was discovered On the other hand, the movement of intertissue ADV-Bs was that several mechanisms of trans-membrane drug delivery exist and that not restricted and the distance per pulse was 48±17 μmfor 5 they are highly dependent on the acoustic parameters, microbubble con- MPa and 88±26 μm for 10 MPa. Cavitation of intertissue ADV-Bs ditions, and the cell-type used. Despite promising results from these study, was observed to split into two bubbles and coalesce back to the advancement from single cell-bubble interactions to clinical use has one during US stimulation (Fig. 1. B). In Fig. 1. C, the intracellu- not been made. This gap in development is largely because the under- lar enhancement of PI indicated cell membrane damage after lying mechanism of trans-membrane drug delivery under high flow condi- ADV-Bs shoving (yellow arrows). The cell and ADV-Bs were tion and for a large population of cells, remains poorly understood and blown out of the original position at 15 s because the cell stuck poorly controlled. Our study explores the ultrasound and microbubble-me- on theADV-Bsand waspushedby pressure gradient of US. diated trans-membrane drug delivery efficiency and safety to a monolayer CONCLUSIONS This study revealed intravital imaging to observe ADV- of endothelial cells using a state-of-theart physiologically-relevant cultiva- B formation, cavitation, and movement in vessels and tissue. Intertissue tion system under different ultrasound exposure conditions. In the end, ADV-Bs could be pushed to distant tissue by US stimulation, and pro- we will evaluate the critical question on whether safe transmembrane duced cell membrane damage when ADV-Bs cavitated or shoved cells drug delivery can be achieved in such a complex, physiologically relevant during movement. Since the treatment resistant tumor cells live away environment. from vessels, these results provided a possibility to touch these cells METHODS Human umbilical vein endothelial cells (HUVEC) were se- and damaged them by ADV-B movement and cavitation. Moreover, the lected as our model cells since they are well described due to their ex- cellular bioeffect induced by physical damage also presented a poten- tensive use in vascular biology and are similar to the arterial endothelial tial way to overcome the resistance of drugs and radiation in tumor cells we’d like to treat in future studies (e.g., in Atherosclerosis). They therapy. were cultured as a monolayer in a flow chamberunder static or pulsatile Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 28 of 122 shear stress (10 dyne/cm2) and at 37°C for 3 days (Fig. 1a, b). A focused ultrasound transducer (0.5 MHz) sonicated the flow chamber in the pres- ence of lipid-shelled microbubbles and during flow conditions (fluid flow rate: 17 ml/min) (Fig. 1c). A centrally-inserted transducer (7.5 MHz) was used to determine the type of microbubble activity generated by pas- sively detecting cavitation emissions during sonication. After sonication, trans-membrane delivery was assessed by quantifying the intracellular uptake of propidium iodide (PI), which is a DNA-binding fluorescent dye Fig. 2 (abstract O43). The effect of acoustic pressure and pulse that is normally impermeable to the cell membrane. Cell viabilitywas duration on (a) Transmembrane delivery and (b) cell viability. The assessed by Calcein-AM. A range of ultrasound parameters (pressure, percentages of transmembrane delivery and cellviability are shown pulse length and exposure duration) and microbubbles conditions were with respect to pressure at 50, 75, 100 and 150 kPa with pulse length systematically evaluated to identify different trans-membrane delivery of 500 and 1000 cycles. Exposure condition: fc=0.5 MHz, PRF=1.4 Hz, and safety profiles. Np =500with microbubbles at 1X clinical dose. The results shown RESULTS The main finding of our study is that safe trans-membrane represents the mean for at least three independent measurements drug delivery can be produced in a clinically targetable cell type (i.e., endothelial cells) underphysiologically relevant conditions (i.e., high fluid velocities) (Fig. 1d). This safe delivery is indicated by cells where red propidium iodide is co-localised with green calcein. A range of O44 ultrasound parameters was explored and we were able to character- Enhanced angiogenesis and perfusion via delivery of basic ise different biological effects: cell viability with and without trans- fibroblast growth factor using an acoustically responsive scaffold 1,2 2 2 2, 1 membrane delivery, cell death and cell detachment. Trans- Alexander Moncion , Melissa Lin , Eric O'Neill , Oliver D. Kripfgans , 3, 4 4 1, 2 membrane delivery increased and cell viability decreased with pres- Renny T. Franceschi , Andrew J. Putnam , Mario L. Fabiilli sure ranging from 50 kPa to 150 kPa, at pulse length of 500 and Applied Physics Program, University of Michigan, Ann Arbor, Michigan, 1000 cycles and in the presence of microbubbles (Fig. 2). It was seen USA; Department of Radiology, University of Michigan, Ann Arbor, that at the highest pressure and pulse length evaluated, transmem- Michigan, USA; School of Dentistry, University of Michigan, Ann Arbor, brane delivery occurred in9.68% ± 10.96% of cells while only 13.18 ± Michigan, USA; Department of Biomedical Engineering, University of 7.33% maintained cell viability. At the lowest acoustic pressure and Michigan, AnnArbor, Michigan, USA pulse length evaluated, trans-membrane delivery occurred in 1.12% Correspondence: Alexander Moncion ± 1.35% of cells while cell viability was 92.45% ± 9.04%. There were Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O44 rapid transitions in cell viability and drug delivery efficiency between 75 kPa and 100 kPa. OBJECTIVES Regenerative growth factors (GFs) are expressed in a CONCLUSIONS This study demonstrated that safe trans-membrane spatially- and temporally-regulated manner during wound healing. drug delivery can be produced in a clinically targetable cell type (i.e., However, delivery of GFs using conventional hydrogel scaffolds endothelial cells) and under physiologically relevant conditions (i.e., does not afford such control, which has hindered the translation high shear stress condition). However, the trans-membrane delivery of GF-based therapies in tissue regeneration. We have developed efficiency was low. This implies that there is a specific microbubble acoustically-responsive scaffolds (ARSs), which are hydrogels dynamic within the exposed tissue volume that can produce the de- doped with sonosensitive perfluorocarbon emulsions containing sired bio-effect, but that it is being produced in only a small subpop- encapsulated GFs. When ARSs are exposed to ultrasound (US) ulation of microbubbles. In future works we will identify this safe and above the threshold for acoustic droplet vaporization (ADV), the effective dynamic and develop pulse sequencing technologies to GF is released from the emulsion. This work studies how ARSs control it so that we can maximise safe drug delivery. can be used to enhance angiogenesis in an in vivo model via the controlled delivery of basic fibroblast growth factor (bFGF), which can stimulate blood vessel formation. METHODS bFGF was encapsulated in monodispersed, perfluoro- hexane (C6F14) double emulsions (mean diameter: 13.4±0.04 μm). ARSs (0.3 mL volume, 10 mg/mLfibrin, 1% (v/v) emulsion) were prepared with 1 μg of bFGF per implant, and injected subcutaneously in the lower back of BALB/c mice. After polymerization, a subset of the implants were exposed to US (2.5 MHz, Pressure = 8 MPa peak negative, 13 cycles, 100 Hz pulse repetition frequency) for 2 min daily. To assess perfusion in and around the implants, the mice were imaged on days 0,1 3, 7, 10, and 14 with a PeriMed Laser Speckle Contrast Imaging (LASCA) system. Blood vessel density in the implants was de- termined via CD31 immunohistochemical staining on days 7 and 14. RESULTS LASCA (Fig. 1): A greater percent change in perfusion, rela- tive to day 0, was observed in ARSs exposed to US (i.e., +US) versus ARSs not exposed to US (i.e., –US) on days 7 (97.0 ± 14.9% vs. 54.0 ± Fig. 1 (abstract O43). Transmembrane drug delivery under flow 13.1%) and 10 (46.9 ± 7.0 vs. 14.3 ± 5.0), respectively. CD31 (Fig. 2): condition. Human umbilical vein endothelial cell (HUVEC) were stained for More blood vessels were present in the +US ARSthan the –US ARS cell viability by calceinAM (green cells) and transmembrane delivery by on day 7 (126.8 ± 23.8 vs. 73.1 ± 21.2 BVs/mm2). Sustained angio- internalization of propidium iodide (red cells). (a) Microbubbles were made genesis was observed with ARSs when compared to fibrin loaded to flow (fluid velocity: 95 mm/s) through a collagentreatedflow chamber with bFGF. with a channel height of 600 μm and a volume of 150 μL. (b) The regioin CONCLUSIONS US enables non-invasive, controlled release of (dotted rectangle) acted as nonsonicated control region, while (c) central bFGF from an ARS, which locally enhanced angiogenesis and per- chamber region acted as sonicated region (solid rectangle). (d) Zoomed in fusion. These finding could be applied in therapeutic angiogen- view at delivery region; a population of cell that exhibited a mixture of esis for the treatment of cardiovascular diseases. Future work will viable cells (green cells) with and without transmembrane delivery and dead attempt to demonstrate efficacy in ischemic pathologies as well cell (red cell) as controlled release of multiple bioactive molecules. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 29 of 122 transducer (3.57 MHz, 0.46x3.55 mm focal area) was used for stimulation while initially exploring the following parametric space: 0.7-5.4 MPa peak negative pressure, 4ms-1s stimulation duration,5-100% dutycycle,0.01- 1kHz PRF.Targetingof the nerve was conducted through a 18.5 MHz, 128-element imaging probe. Once efficacious parameters had been determined (the ability to elicit subsequent EMG responses after an initial re- sponse), the safety of the procedure was explored along with a comparison to electrical stimulation. H&E histology of the tar- geted region (n = 4 stimulation, n = 1 positive control, n = 1 negative control) were employed in a blinded study to detect damage (red blood cell extravasation, inflammation, cell mem- brane rupture). Mice (n = 4 stimulation, n = 4 control) com- pleted an open field behavioral test to explore the short-term safety of the experiment. FUS EMG responses (peak-to-peak, delay to signal) were compared to direct electrical stimulation (1-10 V, 1-10 mA, 200500μs). To measure temperature increase during FUS stimulation, thermocouples were embedded in Fig. 1 (abstract O44). Percent change in perfusion relative to day 0. ex vivo mouse limbs alongside the sciatic nerve and exposed to Peak perfusion was observed on day 7, while the +US condition had FUS stimulation. A force balance was used to determine the statistically greater perfusion both day 7 and 10 acoustic radiation force generated by the transducer to estimate the tissue deformation in the focal region. To confirm neural ac- tivity at the single-unit level, a ex vivo skin-saphenous nerve prep (n = 15) was also stimulated by FUS with the aforemen- tioned parameters while recording extracellular electrophysi- ology responses. RESULTS Non-invasive stimulation of the sciatic nerve via FUS was shown capable of eliciting both observable leg twitching and measurable EMG responses when using the following FUS param- eters: 0.2-5.7 MPa, 35-100% DC (continuous wave), 0.1-1kHz PRF, 0.8- 10.5 ms stimulation duration. Increasing the pressure and DC raised the average peak-to-peak EMG response along with the success rate (Fig. 1). Varying the PRF or stimulation duration did not have a significant ef- fect on response. Both delay and peak-to-peak EMG responses for FUS stimulation were found to be comparable to direct electrical stimula- tion of the sciatic nerve. Mice stimulated with efficacious parameters did not display any significant deviation in behavior compared to the control group or baseline values. The blinded histology study did not detect any damage for the stimulated group, only for the positive con- trol. In ex vivo experiments, FUS was able to stimulate action potentials Fig. 2 (abstract O44). Blood vessel counts determined using CD31 firing in multiple classes of peripheral neurons. Latency to action po- immunohistochemistry. Greater blood vessel formation was observed tential firing was dependent on sensory neuron type and FUS for ARSs with GF exposed to US on day 7 intensity. CONCLUSIONS FUS stimulation of the sciatic nerve was shown cap- able of eliciting physiological responses in vivo. This demonstrates that FUS can be a non-invasive alternative to conventional thera- peutic methods. Specific FUS parameters has been identified for suc- cessful and safe stimulation. Future work to explore the potential mechanisms of generation of the action potential will dictate the O45 FUS parameters to translate this technique to clinical applications. Non-invasive peripheral nerve stimulation via focused ultrasound M. Downs, S.A. Lee, Y. Baba, B. Hoffman, G. Yang, D. Kim, E. Lumpkin, E. Konofagou Columbia University, New York, New York, United States Correspondence: M. Downs Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O45 OBJECTIVES Focused ultrasound (FUS) has been demonstrated to modulate neuronal activity in both the central (CNS) and per- ipheral (PNS) nervous system. While modulation of the CNS has successfully elicited both motor and sensory physiological re- sponses, PNS modulation has only been shown to induce inhibi- tory effects, primarily ex vivo. The ability to non-invasively stimulate and inhibit the PNS with FUS would allow clinicians an alternative therapeutic modality to treat peripheral neur- opathy, as current treatment procedures can generate systemic side effects or require invasive procedures. In this study, we Fig. 1 (abstract O45). EMG responses to FUS stimulation of the aim to show that FUS can elicit excitatory effects targeting the sciatic nerve. The green and red vertical lines are the start and end PNS and generate downstream physiological responses. of FUS stimulation respectively. The blue line is the EMG response to METHODS Focused ultrasound was used to target the sciatic FUS stimulation. Stimulation parameters were as follows: 3.5 MHz, 3.7 nerve in mice while recording EMG responses from the tibialis MPa, continuous wave, 0.88ms stimulation duration anterior muscle in sedated mice (n = 25). A single-element HIFU Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 30 of 122 O46 O48 The safety and feasibility of high intensity focused ultrasound in Clinical experience of intra-operative high intensity focused treatment of resistant hypertension ultrasound in patients with colorectal liver metastases: Results of a P. You Phase II study 1,2 1, 2 2 2 1, 1, 2 State Key Laboratory of Ultrasound Engineering in Medicine Co-founded D. Melodelima ,A.Dupre , Y. Chen , D. Perol , J. Vincenot M. Rivoire 1 2 by Chongqing and the Ministry of Science and Technology,Chongqing Key LabTAU - U1032, INSERM, Lyon, Rhône Alpes, France; Centre Leon Laboratory of Ultrasound in Medicine and Engineering,College of Berard, Lyon, France Biomedical Engineering,Chongqing Medical University, Chongqing, China Correspondence: D. Melodelima Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O46 Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O48 OBJECTIVES To evaluate the safety and feasibility of high intensity This abstract is not included as it has already been published: focused ultrasound (HIFU) in the treatment of resistant hypertension. Melodelima D, Dupre A, Vincenot J, Chen Y, Perol D, Rivoire M. A49 Clin- METHODS 40 patients with resistant hypertension underwent the ical experience of intra-operative High Intensity Focused Ultrasound in treatment of high intensity focused ultrasound, Intraoperative and patients with colorectal liver metastases. Results of a Phase II study. J Ther postoperative adverse effects were recorded; the drop of blood pres- Ultrasound. 2016; 4(Suppl 1):31. Available from: https://jtultra- sure, types of drug used, peak systolic velocity of renal artery and sound.biomedcentral.com/articles/10.1186/s40349-016-0076-5. renal function was followed up to 6 months post treatment. RESULTS (1) All patients completed HIFU treatment successfully.The major discomforts that patients complained during the procedure was O49 pain in the treatment area,which usually disappeared within 24 hours; Catheter-directed thrombolysis of deep vein thrombosis enhanced Base on the SIR classification,most of adverse effects were classified as by intraclot microbubbles and ultrasound: A clinical study A to B, no C to F was found.(2) Preoperative and postoperative peak 1 1 2 1 3 Q. Zhu , S. GAO , G. Dong , M. Guo , F. XIE systolic velocity of left and right renal artery were no statistic differen- 1 Department of Ultasound, XinQiao Hospital,Third Military Medical ce(P=0.635,P=0.688).The comparison between baseline and 1,6 months 2 University, Chongqing, China; Department of Ultrasound, The First of blood urea nitrogen, serum creatinine and glomerular filtration rate Affiliated Hospital of Zhengzhou University, Zhengzhou, Henan, China; were no statistic difference (P=0.772, P=0.652, P=0.366). (3) The systolic 3 Internal Medicine Cardiology, University of Nebraska Medical Center, blood pressure dropped 21.5,23.3,22.4mmHg (P=0.000), diastolic blood Omaha, NE, China pressure dropped 11.1,12.9,12mmHg (P=0.000).24-hour systolic blood Correspondence: Q. Zhu pressure dropped 13.6, 15.2, 14.3mmHg (P=0.000),24-hour diastolic Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O49 blood pressure dropped 5.5,6,4.4mmHg(P=0.000); and types of drug dropped 0.8,0.9,1(P=0.000) in the later 1, 3, 6 months. OBJECTIVES This study is aimed to investigate the safety and effective- CONCLUSIONS HIFU can be safely and feasible used in the treatment ness of treating deep vein thrombosis (DVT) using catheter-directed ther- of resistant hypertension. However, further study about long-term apy (CDT) combined with intraclot microbubble-enhanced ultrasound safety and efficacy of HIFU is still needed. therapy (IMUT). METHODS Fourteen patients with acute DVT (<14 days) undergoing CDT were consented to accept coordinated IMUT treatment as the ex- O47 perimental group. During CDT process, percutaneous therapeutic ultra- Efficacy and influential factors of focused ultrasound therapy for sound (TUS) and trans-catheter injection of SonoVue® microbubbles were NNEDV 1,2 simultaneously performed for about 20-30 minutes twice a day depend- C. Li ing on the length of thrombus. A TUS device (SL-10 Sonolyser, Welld The College of Biomedical Engineering, Chongqing Medical University, Medical Electronics Co., Ltd., China) equipped with a single-element, non- Chongqing, Chongqing, China; Haifu Hospital of the First Hospital focused transducer was used for ultrasound thrombolysis. The transducer Affiliated Hospital, Chongqing Medical University, Chongqing, China was operated at the frequency of 1.0 MHz with the duty factor of 0.01 Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O47 and the peak negative pressure was 750kPa or 1.0 MPa depending on the depth of the thrombus. One vial (5mL) of SonoVue® microbubbles OBJECTIVES To investigate the effectiveness and safety of focused into 15 mL saline was infused constantly into the catheter during the ultrasound for treating non-neoplastic epithelial disorders of vulva treatment. Figure 1 shows the working parameters of the therapeutic (NNEDV) and to analyse the factors that affect the effectiveness of fo- ultrasound device. The time frame could be seen in Fig. 2. The other sixty cused ultrasound. acute DVT patients treated with the same CDT procedure without com- METHODS 136 patients with pathologically confirmed NNEDV under- bining IMUT were retrospectively reviewed for the treatment days and went focused ultrasound treatment. Patients were followed up on a overall urokinase dosage as the control group. The criteria for terminating regular basis after treatment. The efficacy was evaluated based on thrombolysis and extubation were vessel recanalization confirmed by degrees of vulvar itching, physical signs and changes in pathological contrast-enhanced ultrasound (CEUS). The major complications like finding in local lesion. The relation between age, course, status of hemorrhage were monitored. The average treatment days and overall menopause, pathological type and curative was analyzed. urokinase doses of the two groups were compared by Independent Sam- RESULTS The average follow-up period was 23.8 months (range 3 ple Test. months to 60 months). The complete remission occurred in 68 of 136 patients. The cure rate was 50% (68/136). 59 were found much RESULTS Images of a patient in the experiment group were showed more improved. The effective rate was 93.38% (127/136). 9 were inef- in Fig. 3. The average treatment days of the experiment group (4.2 fective and the inefficiency was 6.6% (9/136). 7 cases recurred and ± 1.5 d) were significantly less than that of the control (11.8 ± 4.4 the recurrence rate was 5.51%(7/127). No severe side effects were d) (p<0.01). Also, the overall dosage of urokinase used in the ex- found during treatment and no complications were observed during periment group (3.45 ± 1.66 million IU) dropped significantly about follow-up. The age, course of disease and status of menopause were 28.2% when compared to the control group (4.80 ± 2.47 million IU) related to the efficacy (c2=21.017, P= 0.000; c2=26.591, P= 0.000; (p<0.01). Figure 4 presents the results between the two groups. No c2=8.199, P= 0.000). There was no significant difference in the effi- intracranial and local hemorrhage events happened in both groups. cacy of different pathological types (c2=1.635, P= 0.442). CONCLUSIONS By combining IMUT in CDT treatment of acute CONCLUSIONS NNEDV can be treated with focused ultrasound ef- DVT, the treatment days and overall urokinase dosage were re- fectively and safely. The course of the disease, menopause status and markably reduced. This method may help to short hospital stay the age of the patients can be considered as the predictive factors. and reduce the risk of hemorrhage. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 31 of 122 Fig. 4 (abstract O49). See text for description O50 Influence of "T2-RIM Sign" on immediate therapeutic responses to magnetic resonance-guided high-intensity focused ultrasound ablation of uterine fibroids 1 2, 3 3, 4 3 4, 5 5 S. Yeo , Y. Kim , H. Lim , H. Rhim , S. Jung , N. Hwang Radiology, University Hospital of Cologne, Cologne, Germany; 2 3 Radiology, Mint Hospital, Seoul, Korea; Radiology and Center for Imaging Science, Samsung Medical Center, Seoul, Korea; Health Sciences and Technology, SAIHST, Sungkyunkwan University, Seoul, Korea; Biostatistics and Clinical Epidemiology Center, Research Institute for Future Medicine, Samsung Medical Center, Seoul, Korea Correspondence: S. Yeo Fig. 1 (abstract O49). See text for description Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O50 OBJECTIVES Magnetic resonance-guided high-intensity focused ultrasound (MR-HIFU) is a thermal ablation technique, which is gaining acceptance as an alternative treatment option for pa- tients with symptomatic uterine fibroids. To date, previous studies have shown that different MRI characteristic of the uterine fi- broids, such as high overall signal intensity on T2-weighted MR images and high perfusion, can be correlated with poor thera- peutic outcome. Here, we present a study investigating the influ- ence of a high-signal-intensity peripheral rim on T2-weighted MR images (i.e., T2-rim sign) on the immediate therapeutic response of MR-HIFU ablation of uterine fibroids. METHODS This retrospective study was approved by the institutional review board, and patient informed consent was obtained for MR- Fig. 2 (abstract O49). See text for description HIFU ablation. In total, 196 fibroids (diameter 6.2±2.6 cm, range 3.1- 13.6 cm) in 123 women (age 43.4±5.0 years, range 26-55 years) who underwent MR-HIFU ablation from January 2013 to April 2016 were included. The presence of a T2-rim sign and its corresponding percent coverage around the uterine fibroids were assessed on T2- weighted MR images acquired on the day of the treatment. The effects of a T2-rim sign on the immediate therapeutic responses (non-perfused volume [NPV] ratio, ablation efficiency [NPV/treatment cell volume], ablation quality [grade 1-5, poor to excellent]) were in- vestigated with univariable and multivariable analyses using GEE (generalized estimating equation) analysis. In multivariable analysis, T2 signal intensity ratio of fibroids-to- skeletal muscle, relative peak enhancement of fibroids, and subcutaneous fat thickness were also considered. RESULTS The presence of a T2-rim sign significantly lowered the NPV ratio (54.0±28.0% vs. 83.7±17.7%), ablation efficiency (0.6±0.5 vs. 1.3 ±0.6), ablation quality (3.1±1.2 vs. 4.2±0.8), (P<.0001). There were significant negative correlations between the percent coverage of a T2-rim sign and the NPV ratio (ρ=-0.4648, P<0.0001), ablation effi- ciency (ρ=-0.5086, P<0.0001), and ablation quality (ρ=-0.5086, P<0.0001). GEE analysis showed that the presence of a T2-rim sign and its corresponding percent coverage were independently signifi- cant for ablation efficiency and ablation quality (P<0.05). CONCLUSIONS Uterine fibroids with a T2-rim sign showed signifi- Fig. 3 (abstract O49). See text for description cantly poorer immediate therapeutic responses to MR-HIFU ablation. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 32 of 122 O51 continuously updated heat source position. In order to verify the ac- The current clinical applications of MR guided focused ultrasound curacy of the FUS model, temperature distributions were both mod- surgery in China elled and measured. The experimental arrangement (see Fig. 1) H. Wang consisted of a cell containing collagen gel sandwiched between two Radiology, Shanghai General Hospital, Shanghai, China disks of acoustically absorbing PVA gel, in a water tank. This gel Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O51 sandwich was exposed to heating from the moving focus (0.5mm lat- eral extent, 0.5-1.5kW/cm peak intensity) from a single element OBJECTIVES To demonstrate the history and the current state of clin- focused transducer (Sonic Concepts, Bothell, USA, 1.66MHz). The ical applications of MR guided focused ultrasound surgery (MRgFUS) circular trajectory (6-8mm Ø) was traced out for 300s with a period -1 in China. of 1s . The temperature in the cell layer at the centre of the exposed METHODS The history and the current state of clinical applications of circle was measured using a fine wire thermocouple (RS Instruments, MRgFUS from 2011, when it was first introduced into China, was Corby, UK). Simulations of this arrangement were performed, reviewed and surveyed. together with a sensitivity analysis of the thermal conductivity, RESULTS MRgFUS had been approved for the treatment of uterine fi- specific heat capacity, and attenuation coefficient of the materials broids by the China Food and Drug Administration (CFDA) at 2013. used. Water parameters were assumed for speed of sound, and PVA The other registered clinical study of MRgFUS for CFDA is for the pal- gel density. Using these physical and biological model calibrations, liation of pain in bone metastasis, which also had been completed at the viability distribution in cell layers exposed to FUS mediated HT 2015. Moreover, MRgFUS is in ongoing clinical trials for the treatment will be predicted, and compared qualitatively to an experimental cell of adenomyosis of the uterus, uterine incision pregnancy, and oste- viability (MTT) assay. oid osteoma in some institutions. So far, more than 500 MRgFUS RESULTS For biological modelling, the implementation of reproduct- treatments were completed in China. Some institutions are going to ive rather than immediate cell death was essential for the growth re- start clinical trials for the MRgFUS treatment of facet joint syndrome, sponse dynamics of treated cells to be correctly captured. Having painful knee arthritis, breast adenoma, prostate cancer and tremor. calibrated the growth and death rates in the simulation with those CONCLUSIONS Although it was introduced into clinic in China just of HCT116 cells, more growth curves could correctly be predicted several years,MRgFUS,the new noninvasive thermal ablation method, computationally for homogeneous HT, and/or RT treatments (see Fig. already demonstrates its huge potential and prospect. 2). Using the experimental arrangement described above, thermal doses in the therapeutic range (50-250 CEM) were achieved in the cell layer within the heated circular area. The simulation of the time- temperature curves was in good agreement with experimental data once the physical parameters had been adjusted (see Fig. 3 left). O52 However, experimental variations in the thermal and attenuation A predictive simulation framework for combined focused properties, and set-up uncertainties between different gel samples ultrasound hyperthermia and radiation treatment modelling at a had a strong influence on the time-temperature profiles and thermal cellular level dose (see Fig. 3 right). The corresponding measured material proper- 1 2 1 3 1 S. C. Brueningk , G. G. Powathil , P. Ziegenhein , J. Ijaz , I. Rivens , ties, and qualitative comparisons of simulated and experimental cell 4 1 1 M. Chaplain , U. Oelfke , G. ter Haar viability data of treated cell-containing gels will be presented. Joint Department of Physics, The Institute of Cancer Research, Sutton, CONCLUSIONS This CAM presented can easily be adapted to different 2 3 4 UK; Swansea University, Swansea, UK; Bristol University, Bristol, UK; St. cell lines and treatment scenarios. It therefore has potential for use in Andrew's University, St. Andrews, UK future studies of more realistic, tumour-like cell populations (e.g. spher- Correspondence: S. C. Brueningk oids), and their response to FUS and RT. Such simulations may help to Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O52 identify and optimise treatment schedules & exposure conditions. OBJECTIVES Combined radiotherapy (RT) and hyperthermia (HT) treat- Reference ments offer great potential for the successful treatment of radiation- [1] G. Powathil et al., Semin Cancer Biol, (30, p.13–20), 2015 [2] B. Clarke, resistant tumours by thermo-radio-sensitisation. Focused ultrasound (FUS) Ultrasound Med Biol, (21, p. 353–363), 1995 can be used to induce local HT. For treatment planning, it is essential to quantify the biological effects of such combination treatments. For this purpose, we present a multiscale systems oncology simulation framework. The objectives of this study are (1) to design and simulate in vitro FUS ex- periments, (2) to verify FUS simulation using measured temperature distri- butions, (3) to predict the cellular effects of combination treatments. METHODS A 3D cellular automaton model (CAM), based on the com- putational concepts outlined in [1], was implemented in C++, to model cell populations and their treatment response. Each cell undergoes an individual cycle that regulates its treatment response and proliferation. A separate cell survival model was used to calcu- late surviving fractions for the combinations of radiation and thermal doses delivered. From this, a known proportion of cells would undergo immediate, or reproductive, cell death via mitotic catastro- phe. The CAM was compared to results from experiments designed to characterise the response of HCT116 cells in vitro. Clonogenic as- says, cellular growth curves, and flow cytometry analysis were used Fig. 1 (abstract O52). Experimental arrangement used. A gel sandwich to assess overall survival, growth dynamics, and cycle distribution consisting of a thin (~300μm thick), cell containing collagen gel and two after homogeneous heating and/or irradiation. In particular, the influ- 6mm thick slices of PVA cryo-gel is enclosed in a sterile sample holder ence of the cell kill model used in the simulation (instantaneous vs. filled with degassed cell culture medium. Top and bottom of the sample reproductive cell death) was tested. A linear propagation model of holder were sealed with tight Mylar membranes to minimize beam FUS exposure [2] was implemented. To achieve a heated volume sig- reflection. The sample holder is placed with the cell layer located at nificantly larger than the geometric focus, a transducer was moved the focal plane of a focused single element transducer in a water tank. at constant speed in a circular trajectory. The majority of the heated For treatments, the transducer is moved on a circular trajectory with cells are not exposed to FUS directly, but are heated by diffusion of temperature being monitored by a fine wire thermocouple at the thermal energy into the circle’s centre. Heat generation and diffusion center of the circle at the cell layer were simulated by iterative solution of the bioheat equation with a Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 33 of 122 and are different in their density and Young's modulus. The Boundary Element Method is employed to solve the potential flow surrounding the bubble. The bubble is assumed to be adia- batic and initially stable. RESULTS It is seen in Fig. 1 (a) that an oscillating bubble col- lapses with a jet towards the elastic membrane (Sankin & Zhong, 2006). Our simulation result (Fig. 1(b)) compares well to the ex- perimental observationinbubbleshapes and thetimingof events. Figure 2 shows the collapse of the bubble near the elastic membrane after being hit by a shock wave at 114 μs. It is observed in both experiment and simulation that the bubble splits as the membrane moves towards the bubble. We use our Fig. 2 (abstract O52). Comparison of experimental and simulated model to study the effect of initial bubble size in the bubble growth curves for HCT116 cells treated with different modalities. shock waves interaction near fat tissue. It is seen in Fig. 3(a) that Left: 2Gy radiation (22000 cells seeded in 24-well plates, surviving the small bubble of 100 μm in radius collapses with a high speed fraction S2Gy = 0.43(0.38,0.49)). Right: Combination of 2Gy radiation jet towards the fat tissue (and the fat tissue moves towards the and heating for 5min at 46°C (300000 cells seeded in 6-well plates, collapsing bubble). However, if a larger bubble is present (in this S2Gy+5min,46C = 0.11(0.06,0.19)). Cells were simulated with a mean case, 500 μm in radius), the bubble may split before collapsing doubling time of 19.5h. The experimental data points (black circles) (Fig. 3(b)). The fat tissue also moves closer to the bubble just be- are shown, as are the simulated total cell numbers (solid lines), and fore it collapses. We have also examined the effect of shock simulation of the 95% confidence bounds of the surviving fractions wave reflection at rigid interfaces such as bone. Figure 4(a) used (dashed lines) shows the collapse of a 10 μm near bone tissues (on top) after being hit by the lithotripter shock wave. The shock wave is reflected with 0.54 of the original amplitude at the bone inter- face. The bubble collapses at 0.148 μs. Figure 4(b) shows the simulation results without the consideration of the shock wave reflection. The bubble collapses only at 0.166 μs. Thesizeofthe jet is also significantly larger in this case. CONCLUSIONS The interaction between a stationary bubble near various bio-materials and a lithotripter shock wave has been suc- cessfully modeled and simulated using the Boundary Element Method. High speed jets are developed in the bubbles in the direction of travel of the shock waves as previously reported in literature. It is therefore concluded that presence of the bio- Fig. 3 (abstract O52). Comparison of experimental and simulated materials causes the bubble to collapse faster and with slightly time-temperature curves for moving focus (1143 W/cm peak intensity, higher jet speed. However, when the initial bubble size is varied, 6mm diameter of the circular trajectory). Left: The logged temperature different bubble dynamics is observed. In the case of very large data (blue) is first smoothed using a moving average (red), and then bubble (500 μm in radius), the bubble does not collapse with a -1 -1 simulated (yellow) using a thermal conductivity of 0.49Wm K ,aspecific jet towards the bio-material, but will split into smaller bubbles. -1 -1 heat capacity of 2000Jkg K , and an acoustic absorption coefficient of The reflection of the shock wave near rigid boundary such as 0.49nepers/cm at 1.66MHz. Right: Comparison of the simulation (solid bone or stone has the effect of causing the bubble to collapse yellow line) with repeat measurements in the same experimental set-up faster with a narrower jet. for different PVA/gel samples (smoothed data, dashed lines) illustrating experimental uncertainties in thermal dose (15-142CEM) Reference Sankin, G. N. & Zhong, P. (2006), ‘Interaction between shock wave and single inertial bubbles near an elastic boundary’, Phys. Rev. E 74, 046304. O53 Lithotripter shock wave interaction with a bubble near various bio-material 1 1 2 3 S. Ohl , E. Klaseboer , A. Szeri , B. Khoo Institute of High Performance Computing, Singapore, Singapore; 2 3 University of California, Berkeley, Berkeley, California, USA; National University of Singapore, Singapore, Singapore Correspondence: S. Ohl Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O53 OBJECTIVES The interaction of a lithotripter shockwaves with a gas bubble which is located near bio-materials such as fat, skin, muscle, cornea, cartilage, and bone is studied using numerical simulation. Our model is verified with comparison to experimen- tal observations from Sankin & Zhong (2006) where a bubble col- lapses near an elastic membrane after it interacts with a lithotripter shock wave. We proceed to perform a systematic study of the various factors influencing the bubble collapse, in- cluding wave profile, initial bubble size, and reflection of the shock waves at rigid interface (such as bone). The results may be Fig. 1 (abstract O53). Interaction of an oscillating bubble near an useful in the design of lithotripster shock wave therapies and the elastic membrane. (a) Experimental results from Sankin & Zhong prevention of collateral damages in medical treatments. (2006). (b) Numerical simulation of the bubble collapse. Both bubble METHODS The shock waves are modeled as a traveling pressure shapes and timing of events compare well with experiment wave across the domain. The bio-materials have linear elasticity Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 34 of 122 O54 Three-dimensional passive acoustic localization and mapping for cavitation: a preliminary study S. Lu, X. Du, M. Wan Biomedical Engineering, Xi’an Jiaotong University, Xi’an, China Correspondence: S. Lu Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O54 OBJECTIVES Passive acoustic mapping (PAM) based on a linear array has been widely applied in real-time monitoring of high intensity focused ultrasound (HIFU) therapy. By applying a beamforming algorithm to the passively received acoustic emis- sion of cavitation, the obtained spatial distribution of cavitation can be used for prediction of HIFU lesions. However, two- dimensional (2D) PA. M can only provide a plane distribution of cavitation, which is not conducive to clinical diagnosis. PAM based on a hemispherical array Fig. 2 (abstract O53). The interaction of a laser-generated bubble has been used for three-dimensional (3D) vascular imaging with high near an elastic membrane after being hit by a lithotripter shock resolution in the brain, but it is not suitable for treatment monitoring wave at 114 μs. The numerical simulation results are overlapped on of other biological tissues, such as liver and kidney. This means that top of the experimental observations from Sankin & Zhong (2006). It 3D PAM based on an area array for omnibearing monitoring of ultra- is seen that in both simulation and experiment, the bubble splits as sound therapy is required. The objective of this work is to develop a the membrane moves towards the bubble three-dimensional super-resolution passive imaging technique for microvessel and an omnibearing monitoring of ultrasound therapy in real time. METHODS A multi-bubble model was used to create acoustic emissions of cavitation source for single bubble and multiple bubbles. The emissions were recorded by a 32 × 32 area array with an aperture size of 38.4 × 38.4 mm. For three-dimensional (3D) localization of single cavitation bubble, the differential times of arrival between elements at various positions and a reference element was firstly calculated by using cross- correlations. Then a paraboloid function derived by Fresnel ap- proximation was used to fit time-delayed curved surface. Through least square fitting, the estimated paraboloid coeffi- cients were used to calculate 3D position of single cavitation Fig. 3 (abstract O53). A gas bubble collapses near fat tissue (on top) source. For passive mapping of extended cavitation region, a when it is hit by a lithotripter shock wave from below. (a) The initial 3D passive beamforming algorithm based on time exposure bubble size is 100 μm in radius. The bubble collapses with a jet acoustic (TEA) algorithm was applied to the 3D prebeamformed towards the fat interface. The fat tissue moves towards the bubble. (b) data to generate 3D cavitation images. In the algorithm, the The initial bubble radius is 500 μm. The bubble may split before source energy at each imaging location was calculated by inte- collapsing. The fat tissue moves more towards this bigger bubble grating the square of the source strength (delay-and-sum of the prebeamformed data) over a time interval. RESULTS Using a paraboloid to fit time-delayed curved surface can accurately localize single cavitation bubble at different po- sitions in three-dimensional (3D) space, which can be used for super-resolution passive imaging of microvesssel. 3D Cavita- tiom images of single bubble at (0 mm, 0 mm, 40 mm) has a full-width at half-maximum of 0.27mm × 0.27 mm × 2.03mm, which can be regarded as the point spread function at a fixed position of single cavitation source and given parameters of area array (Fig. 1). 3D cavitatiom images and its cross sections along three axes for multiple bubbles (distributed spatially with a normal distribution, the standard devi- ation of the distribution is 0.5 mm × 0.5 mm laterally and 1 mm axially) with different number (N = 20 and 50) demonstrated the feasibility of using 3D TEA-PAM to map extended cavitation region (Fig. 2). The “X-type” ar- Fig. 4 (abstract O53). The effect of the shock waves reflection on tifacts in 3D cavitation images were caused by multiple bub- the collapse of a bubble near bone tissue. The 10 μm bubble is bles interfering with each other, there were no actual bubbles located 10.5 μm from the bone interface. (a) The reflected shock in the artifact region. The artifacts were needed to be im- wave is implemented as a downwards traveling wave 0.54 proved by other technique such as adaptive beamformer. And amplitude of the original lithotripter shock wave after it hits the it should be noted that the lateral resolution of 3D cavitation bone interface. The corresponding time for each bubble shape images is significantly better than axial resolution (Figs. 3, 4). (from outer to inner) is 0, 0.123, 0.137, 0.142, 0.146, and 0.148 μs. The results can assist in real-time 3D monitoring of ultrasound (b) Simulation results without shock wave reflection. The bubble therapy. collapses only at 0.166 μs. For the dotted line bubble shapes, time CONCLUSIONS 3D PAM based on TEA has the potential of providing from outer to inner are 0.154, 0.160, 0.166 μs a novel method for 3D real-time monitoring of ultrasound therapy. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 35 of 122 Fig. 3 (abstract O54). The slices of cavitation images using 3D TEA-PAM algorithms along (a) x axis, (b) y axis, and (c) z axis for multiple bubbles. The cross sections of cavitation images along (d) x axis, (e) y axis, and (f) z axis. The nominal distance from cavitation source to the area array is 40 mm. The number of cavitation bubble is 20, distributed spatially with a normal distribution Fig. 1 (abstract O54). Time delay of the 3D prebeamformed data from single cavitation bubble and the corresponding fitting curved surface. The position of bubble is (a) (0 mm, 0 mm, 40 mm), (b) (0 mm, 0 mm, 80 mm), (c) (9.6 mm, 0 mm, 40 mm), (d) (0 mm, 9.6 mm, 40 mm), (e) (0 mm, -9.6 mm, 40 mm), and (f) (-9.6 mm, 0 mm, 40 mm), respectively Fig. 4 (abstract O54). The slices of cavitation images using 3D TEA-PAM algorithms along (a) x axis, (b) y axis, and (c) z axis for multiple bubbles. The cross sections of cavitation images along (d) x axis, (e) y axis, and (f) z axis. The nominal distance from cavitation source to the area array is 40 mm. The number of cavitation bubble is 50, distributed spatially with a normal distribution O55 Towards a robust and general ultrasound propagation model for MR HIFU Tumour treatments: the hybrid ray tracer approach 1 2 3 4 3 D. Modena , L. Sebeke , M. Baragona , A. Elevelt , R. Maessen , 1 1 D. Bošnački , H. Ten Eikelder Eindhoven University of Technology, Eindhoven, Netherlands; 2 3 University Hospital Cologne, Cologne, Germany; Department of Multiphysics & Optics, Philips Research Eindhoven, Eindhoven, Netherlands; Department of Oncology Solutions, Philips Research Eindhoven, Eindhoven, Netherlands Correspondence: D. Modena Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O55 OBJECTIVES Magnetic Resonance-guided high intensity focus ultra- sound (MR HIFU) has shown to have a clinical relevance for the treat- ment of various types of tumours, such as uterine fibroids and bone Fig. 2 (abstract O54). The slices of cavitation images using 3D metastases. MR imaging is used to plan and evaluate the HIFU treat- TEA-PAM algorithms along (a) x axis, (b) y axis, and (c) z axis for ment, whereas MR thermometry using proton resonance frequency single bubble. The position of bubble is (0 mm, 0 mm, 40 mm) (PRF) is used to monitor the procedure. There are different reasons Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 36 of 122 why the development of ultrasound propagation models for MR- HIFU treatment is of interest. Particularly, they allow the improve- ment of pre-operative therapy planning by predicting the location and the amount of energy deposition during treatment as input for temperature evolution calculation. Furthermore, models can over- come the limitations of the PRF thermometry data, which can be ac- quired only in soft tissues and which present a low spatial resolution compared to the dimension of the treatment region. In addition, these models can enable model- based control during treatment. General accepted methods for HIFU propagation are the Far Field Approximation (FFA) and the Angular Spectrum Plane Wave method (ASPW). However, the first mentioned method is valid only when one homogeneous medium is present, and the second one is not applic- able with no-parallel interfaces. Therefore, in this work we present Fig. 1 (abstract O55). The hybrid ray tracer. In green the rays till the and validate a ray tracer method, a fast and flexible (easily adaptable first interface, and in blue the refracted rays in the tissue- mimicking gel to complex geometries, for both soft tissues and bone cases) ap- proach, which will be used as a baseline for an extension towards a patient-personalized HIFU propagation model. METHODS Firstly, the hybrid stochastic ray tracer is presented (see Fig. 1). The model is suitable for the geometry of the 256-element phased array transducer (focal length 14 cm, frequency 1.2 MHz) used in the MR-HIFU system (Sonalleve V2, Philips, Vantaa, Finland). In the model, ultrasound propagation is represented by rays of inten- sities. The word ‘hybrid’ highlights the fact that the rays leave each transducer element in random directions with an initial intensity de- rived from the FFA formula. The rays follow the laws of refraction and reflection at an interface between two different media. The inter- ference of ultrasound waves in the focal region is calculated from the phase assigned to each ray and the output of the model is the produced power density, which is represented by a 3D table with 0.2 mm grid size. Secondly, our model is compared with the ASPW method in terms of power produced in the focal region, in a config- uration with two propagation media (lossless oil and tissue mimick- ing gel) with only flat surfaces (see Fig. 2). Thirdly, experimental validation data are acquired. In the experimental set-up the ultra- sound waves travel through the lossless material and are focused in Fig. 2 (abstract O55). The ASPW and the hybrid ray tracer outputs the tissue mimicking gel containing Zerdine (CIRS Model 054GS Gen- are compared in a configuration with two homogeneous materials, eral Purpose Ultrasound Phantom). From the PRF thermometry, the separated by a flat interface temperature increases in the coronal slice crossing the focal point are recorded (see Fig. 3). Finally, for this configuration the hybrid ray tracer and the ASPW are used to compute the power density, which then functions as source term in the heat equation. The solution of this equation, found by a finite element approach using COMSOL Multhiphysics (COMSOL, Inc., Burlington, MA), can be compared with the experimental data. However, since the thermal parameters of the tissue- mimicking gel are unknown, these parameters are first found by fitting the results of the ray tracer/ASPW-heat equation to the experimental data. In particular, the diffusion coefficient D=λ /(ρc )(where λ stands for t p t the thermal conductivity, ρ the density, c the specific heat at constant pressure) has been fitted by considering temperature data from the cool- ing down phase, and the term β=b/(ρc ) (where b represents the ab- sorption coefficient, that is the fraction of power produced which contributes to the temperature increase) has been fitted taking into ac- count the heating up phase. RESULTS The hybrid ray tracer predicts a power density in good agreement with the reference method ASPW. After fitting the ther- mal parameters, the resulting model temperature prediction are in accordance with the experimental data. Moreover, the thermal pa- rameters found by the fitting, are in the expected range. CONCLUSIONS The hybrid ray tracer is a good starting point to model MR HIFU treatments, it has been developed to overcome the limitations of the other methods for ultrasound propagation, in particular the prob- lematic treatment of no-flat surfaces. Moreover, through the computa- tion of shear and longitudinal waves in the solid material, the method can be applied to the case of the treatment of bone metastases. Ultim- ately, the hybrid ray tracer discloses advantages such as flexibility and high calculation speed, which make this model a suitable first approach Fig. 3 (abstract O55). The coronal planes and the sagittal plane towards a patient- specific and general HIFU propagation model. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 37 of 122 O56 predict the time-varying ultrasound field generated by HIFU trans- Experimental validation of full-wave HIFU simulations in ducers in heterogeneous, absorbing media under both linear and heterogeneous media nonlinear conditions. These models are likely to find many applica- E. Martin, J. Robertson, B. Treeby tions, from simple in silico investigations, through to patient specific Medical Physics and Biomedical Engineering, University College London, treatment planning. It is important to note, however, the close agree- London, UK ment relies on accurate knowledge of the geometry, position, and Correspondence: E. Martin material properties of the heterogeneities in the beam path. While Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O56 this is possible in a laboratory situation, the sensitivity of the simula- tion output to these parameters is not yet fully understood. Further OBJECTIVES Numerical simulations have a wide variety of applica- work is needed to clarify the uncertainties in numerical predictions tions in high-intensity focused ultrasound (HIFU), including trans- using real biological tissue where the geometry and material proper- ducer design, patient selection, treatment verification, and treatment ties may not be precisely known. planning. However, for these simulations to be clinically useful, it is crucial that the accuracy of the numerical model is rigorously estab- lished under realistic conditions. The aim of this study was to quanti- tatively validate the wave models within the open-source k- Wave O57 Toolbox using a series of experimental measurements made with Translating microbubbles with millisecond scale ultrasound pulses: heterogeneous fluid and solid objects in the beam path. implications for controlled transport of bubbles to a boundary 1,2 2 1,2 METHODS Experiments were performed in a 60 x 60 x 100 cm tank C. Acconcia , A. Wright , D. Goertz 1 2 of temperature controlled, degassed, and deionised water. The University of Toronto, Toronto, Ontario, Canada; Sunnybrook Research acoustic field was generated by a single-element spherically focused Institute, Toronto, Ontario, Canada Correspondence: C. Acconcia HIFU transducer driven at 1.1 MHz with a 4 cycle burst. Signals were acquired using a calibrated 75 μm needle hydrophone connected to Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O57 a 5- axis computer controlled positioning system. Planar scans were acquired both parallel and perpendicular to the beam axis following OBJECTIVES To elicit bio-effects, ultrasound (US) stimulated micro- bubbles (MBs) must be in close proximity to their target. In a situ- propagation through the heterogeneity. Measurements were made at several drive levels to give both linear and nonlinear conditions. ation such as sonothrombolysis, however, the majority of circulating Two types of medium inclusions were used to obstruct the beam MBs will not be in proximity to the clot boundary. Radiation forces can path. The first was a series of fluid filled containers (rectangular, potentially direct MBs to clot surfaces and recent work (Acconcia et al., wedge, and planoconvex) constructed from laser cut Perspex and a JASA 2016) has suggested clot degradation patterns (Fig. 1A) could be stretched Mylar membrane. These were filled with either olive oil or influenced by ‘transport pulses’. However, this is a complex process in- glycerol, and were intended for validation of the models within a volving a population of MBs flowing in a vessel, with size-dependent ra- soft-tissue like medium. The second was a series of 3D printed and diation and drag forces. We recently conducted experiments to resin cast skull bone phantoms derived from a T1 MR image. These investigate the process of bubble accumulation at a boundary with a were intended for validation of the models when heterogeneities view to improving exposure strategies in applications such as sono- with a larger acoustic impedance mismatch are present. The trans- thrombolysis. The size and spatial distribution of bubbles arriving at a ducer and medium inclusions were rigidly positioned using a series fibrin clot boundary (Fig. 1B) were found to be highly dependent on of custom-made Perspex and 3D printed mounts and opto- pressure and flow conditions. The ability to model and control this mechanical components attached to a breadboard mounted over process is limited by a paucity of data on the translational dynamics of the water tank. This allowed accurate registration of the source and encapsulated MBs under the influence of millisecond scale US pulses. object positions. For each experiment, a numerical simulation was To date, experimental work for individual MBs has focused on the use conducted using the MPI version of the open-source k-Wave Toolbox of shorter (imaging) pulses. The purpose of this work is to directly in- running on the IT4I Salomon supercomputer. The grid parameters vestigate the translation of individual MBs with millisecond pulses in used 6 spatial points per minimum wavelength for the fluid objects, order to constrain theoretical modeling of this process. and 10 points for the solid objects. The source conditions were METHODS In house developed optical tweezers integrated with a established using free- field measurements and linear acoustic holog- microscope (60x) and high speed camera were employed to select indi- raphy. For measurements at other drive levels, the source pressure vidual Definity MBs and position them in a fluid region away from was assumed to scale linearly with drive voltage. The medium prop- boundaries. MBs were then subjected to a series of 1 ms length, 1 MHz erties within the simulation were specified according to the known pulses and were recorded (10 kframes/s) while translating laterally position and geometry of the scattering object, and using book through the optical field of view. Image analysis was employed to values for the material properties. The measured and simulated wave quantify the displacements as a function of time with the primary fields were compared using several metrics, including the peak posi- metric of interest being the distance travelled within the first pulse. Ex- tive and negative focal pressure, focal volume, focal position, arrival posures were conducted at transmit pressures of 25, 50, 100, 150 and time of reflections, and L2 error in the spatially varying wave field. 200 kPa. For each pressure a range of bubble sizes (1.5-9 microns in For nonlinear conditions, the fields after spectral decomposition were diameter) were assessed (n=86). The radial oscillations of encapsulated also compared. MBs were calculated based on a variant of the modified form of the RESULTS The simulations and experiments showed close quantitative Rayleigh-Plesset equation proposed by Marmottant et al (JASA 2005). agreement. Errors were typically < 3% for the peak positive pressure, This model requires as inputs estimates of shell elasticity, dilitational < 3% for the focal volume, < 1 mm for the focal position, < T/12 for viscosity, and a starting effective surface tension. To assess displace- the arrival times, and < 6% for the L2 error. The exception was the ment, radiation force along with other relevant forces on the MB measurements made using the resin cast bone phantom, where sim- (added mass, quasi-steady drag and history force) were solved numer- ulations overestimated the focal volume by ~12%. This discrepancy ically. The history force integral was evaluated using methods pre- is likely due to variations in the material properties of the resin from sented by Garbin et al. (2009), and Chung et al. (1982). After empirical the casting process, which were not captured by the numerical modification, the history force integral has been generalized to apply model. For the nonlinear conditions, errors were smallest at the over a larger range of Reynolds numbers. For this finite Reynolds num- fundamental frequency, but still remained acceptable at the fifth ber form of the history force (Takemura et al, JFM 2004), we used the harmonic (the highest harmonic measured with significant signal- numerical integration scheme proposed by Chung et al. (1982). to-noise). RESULTS The displacement as a function of MB size and pressure are CONCLUSIONS The results demonstrate that the full-wave models in shown in Fig. 1C. A pronounced feature of the displacement curves the open-source k-Wave toolbox can accurately and quantitatively is the presence of an effective threshold point in size, below which Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 38 of 122 there is only minimal translation. The threshold size increases with OBJECTIVES To reduce hyperthermia treatment times, there has been decreasing pressure. Note that single pulse displacements higher a tendency to operate HIFU in short bursts, at high power levels. This than 140 microns were not observed due to field of view constraints. necessitates the introduction of laboratory thermometry methods that The threshold dependant nature of the onset of rapid displacement are temporally and spatially resolved, and are able to characterize the could be well accounted for by the threshold dependant nature of burst of thermal output in real time. In this work we describe the devel- oscillations captured by the encapsulated MB model. Example mod- opment of a novel optical imaging method for temperature measure- eling is shown for the 25 kPa case (Fig. 1D), with a local maximum ments in a HIFU field that is fast and spatially resolved. The method is about a peak size, which correlates with the approximate resonant non invasive and only requires optical access to operate. size of Definity at 1 MHz. The model results for up to 100 kPa were METHODS The described thermometry method is based on the found to be in good agreement with data, when employing shell pa- temperature-dependence of the optical refractive index in solids, rameters that were within the range of those previously reported for gels, and liquids. As an example, water and polyachrylamide gel have -4 phospholipid agents. Importantly, the inclusion of history force was a linear dependence of around k=-1x10 /°C. When a laser ray travels required to accurately fit the data. through an axisymmetric HIFU heated spot, it will deflect from its CONCLUSIONS This study reports the first size dependant data set straight unperturbed path. The net angular ray deflection is for the translation of MBs under the influence of ms scale US pulses measured and converted to temperature using a direct formula at therapeutically relevant pressures. A key feature was threshold size derived from solving the paraxial eikonal ray equation for a Gaussian dependent displacements, where the degree of translation was phase object. In the experimental setup, a 65 mm aperture HIFU highly sensitive to the radial asymmetry of the oscillations inherent transducer (Sonic Concepts) running at its third harmonic of 1.6 MHz with this model, a behavior that is characterized by compression was focused on an optically transparent tissue mimicking phantom dominated oscillations for smaller bubbles (<~3-4 microns). The ma- cube of 2 cm side length placed in a water bath for acoustic jority of simulations to date have excluded history forces on the basis coupling. A laser light sheet of 3 cm height was passed through a of Reynolds number arguments, however for the conditions investi- custom made comb to chop it into individual rays producing a gated here the inclusion of history force was found to be important. planar bundle formation. The ray bundle was then focused with a When this constrained MB model was applied to our previously ac- cylindrical lens to allow as many rays as possible to pass through the quired data set of size dependant MB arrivals at a planar surface, 0.3 mm heated spot to improve the spatial resolution. The acoustic good agreement was found, thereby providing a tool to investigate and optical axes were orthogonal. The ray bundle was imaged at 200 and optimize the control of translating MBs to a surface. fps, at a location around 25 cm down the laser path from the HIFU spot (providing geometric advantage for imaging) during HIFU irradiation to capture the heating and cooling phases. The ray deflection angles from the individual images (relative to no HIFU image) were extracted and converted into radial temperature profiles. The tissue mimicking phantom cube was made by casting liquid silicone that cured in 24 hrs. K- type thermocouples (TC) of 0.13 mm dia. were cast within the phantom to read temperature. Figure 1 shows a picture of the HIFU installed in the experimental setup and a sample raw image of the ray bundle at the imaging location. Geometric triangulation was used to map the rays from the imaging plane to the HIFU focal plane. RESULTS By comparing sequential ray images, ray deflection maps can be extracted as shown in Fig. 2. The resulting radial distribution of the angular deflection profile at the HIFU focal plane is also shown in the figure. The radial deflection profile is directly substituted in the aforementioed equation (shown in Fig. 2) to produce radial temperature profiels at different time steps as shown in Fig. 3. In this figure, the HIFU was operated to produce a 2 ms burst at a power level of around 20 W. As noted, the optical imaging thermometry is able to resolve the temperature distribution across the 2 mm HIFU heated spot giving a spatial resolution better than 0.1 mm. The temperature profile is narrow with a high peak right after HIFU ex- posure. The profile widens and cools off with time due to heat diffu- sion. In Fig. 4 the temporal evolution of the peak in the temperature distribution is plotted alongside the TC reading. The TC signal spikes Fig. 1 (abstract O57). A) Two-photon microscopy of the erosion higher than the optical ray temperature signal, expectedly due to vis- zone of the flourescently tagged fibrin network of a treated blood cous heating. The two curves agree well during the latter cooling clot. Upon arrival, MBs penetrate and disrupt the fibrin network. B) Top phase. The temporal resolution of the method is fixed by the frame view of the size dependant arrival of MBs at a planar fibrin clot boundary. rate of the camera which in this case is 0.02 ms. Faster frame rates A minimum intensity projection is shown over 100 US pulses. C) Size are straightforward to obtain, but in this case were limited by the and pressure dependant displacement of MBs from a single 1 ms available laser power illumination. pulse. D) A comparison of data and modeling for the 25 kpa case CONCLUSIONS A fast rise-time and spatially resolved optical imaging ray-bundle thermometry method has been developed and demon- strated with milliseconds long HIFU bursts in a tissue mimicking phantom. Unlike previously proposed thermometry methods such as O58 laser-induced fluorescence which mainly works in liquids, the current A fast non-invasive optical imaging thermometry method for HIFU ray bundle method works equally well in liquids and solid gels and it G. Oweis, H. Daoud does not require the tedious calibration steps. It requires working Mechanical Eng, American University of Beirut, Beirut, Lebanon with optically transparent materials. The simplicity of the experimen- Correspondence: G. Oweis tal setup and ease of image processing, in combination with the Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O58 Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 39 of 122 availability of lasers and cameras in most laboratories makes this method a viable choice for fast and resolved characterization of HIFU thermal outputs. Fig. 1 (abstract O58). Experimental setup showing the HIFU in the water bath, the focused laser bundle illuminating the tissue phantom, and imager (left); and a sample raw ray bundle image in the imaging plane (right) Fig. 4 (abstract O58). Temporal evolution of the peak temperature at the center of the heated spot using the optical ray method (black) alongside thermocouple readings (green) O59 In vivo and ex vivo monitoring of thermal ablation in a porcine model using ultrasonic nakagami imaging S. Zhang, S. Shang, Y. Han, R. Xu, C. Gu, L. Zhang, M. Wan Department of Biomedical Engineering, Xi'an Jiaotong University, Xi'an, Shaanxi, China Correspondence: S. Zhang Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O59 Fig. 2 (abstract O58). Two dimensional ray deflection map in the imaging plane (left); and the corresponding radial distribution (r) of OBJECTIVES The uses of high-intensity focused ultrasound (HIFU) the angular ray deflections in the HIFU focal plane (right); also and microwave ablation (MWA) as a non- invasive or minimally inva- shown is the temperature conversion equation sive therapeutic technique are being investigated due to the devel- opment of the monitoring imaging techniques. The acoustic posterior shadowing effects of cavitation and/or boiling bubbles influence the accuracy for defining the location of thermal lesions when using ultrasonic monitoring imaging. This in vivo and ex vivo study investigated the feasibility of using ultrasonic Nakagami imaging to evaluate the thermal lesions during HIFU and MWA in a porcine model. METHODS 2-D RF data backscattered from the ablated region were captured by a modified diagnostic ultrasound scanner to estimate ultrasonic Nakagami parameters of the thermal lesions, and to recon- struct the ultrasonic B-mode and Nakagami images during HIFU and MWA. A term contrast-to-noise ratio (CNR) between the thermal le- sions and the surrounding normal tissue is used to estimate the con- trast resolution of the ultrasonic B-mode and ultrasonic parameter images. RESULTS Unlike Ultrasonic B-mode images, Nakagami images were less affected by the shadow effect in monitoring of thermal ablation, and a fairly complete hyper-echoic region was observed in the Naka- gami image. After thermal ablation, a bright hyper-echoic region ap- peared in ultrasonic Nakagami parameter images as an indicator of the thermal lesion. Mean values of the Nakagami parameter in the thermal lesion region increased to 0.58, 0.71 and 0.91 after 1, 3 and 5 min of thermal ablation. CNR values calculated for Nakagami par- ameter images increased from 0.13 to approximately 0.61 during thermal ablation and then decreased to 0.26 at the end of the post- Fig. 3 (abstract O58). Radial temperature distribution of the HIFU ablation stage. The corresponding CNR values calculated for the heated spot at increasing times from the HIFU pulse ultrasonic B-mode images were 0.24, 0.42 and 0.17. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 40 of 122 CONCLUSIONS This in vivo and ex vivo study on a porcine model suggested that the Nakagami parameter may have the potential use to evaluate the formation of thermal lesions and the ultrasonic Naka- gami imaging may provide an alternative modality for monitoring HIFU and MWA treatment. O60 Changes in the optical scattering and absorption spectra of ex-vivo chicken breast tissue following exposure to HIFU J.L. Raymond, R. Cleveland, R.A. Roy Department of Engineering Science, University of Oxford, Oxford, UK Correspondence: J.L. Raymond Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O60 Fig. 1 (abstract O60). Optical reduced scattering coefficient (μ ') OBJECTIVES Real-time acousto-optic (AO) sensing has been shown versus optical wavelength (λ) for 10 s HIFU exposures at 150, 200 to non-invasively detect changes in ex vivo tissue optical properties and 250 W/cm during high-intensity focused ultrasound (HIFU) exposures. Baseline changes in optical properties have been previously measured as a function of thermal dose for chicken breast exposed to a temperature- controlled water bath (doi:10.1088/0031-9155/59/13/ 3249). In this work, the wavelength-dependent optical scattering and absorption coefficients of ex vivo chicken breast tissue exposed to HIFU were measured using an integrating sphere spectrophotometric technique. METHODS Thin tissue sections (approximately 2 mm) were mounted on an acoustically transparent membrane such that the bottom sur- face was coupled to a 37°C water bath and the top surface exposed to air to permit non-contact thermal measurements using an infrared camera. Thermal damage was induced using a focused 1.1-MHz transducer (H-102; Sonic Concepts, Bothell, WA, USA) coupled to an acoustic lens and positioned in the water bath below the tissue sam- ple. Phase-shifts produced by the lens de-focused the beam and re- sulted in an annular focal zone with the acoustic intensity maximum located 1 mm off-axis. Thus, a larger focal heating area could be pro- duced in the tissue sample with less spatial variation in temperature (and thermal dose) in the region-of-interest (ROI) than for a tightly focused beam. Spatiotemporal surface temperature elevations were measured using an infrared camera (FLIR Systems, Kent, UK) and used to calculate the spatially-dependent thermal dose delivered to the tissue ROI. Optical property changes in the ROI were measured using a dual-beam UV-Vis-NIR spectrometer (Lambda 750s; PerkinEl- mer, Beaconsfield, UK) equipped with a 100 mm integrating sphere. The exposure intensity and time were varied in order to determine the optical property changes in tissue as a function of the delivered thermal dose. RESULTS Figure 1 plots the optical reduced scattering coefficient Fig. 2 (abstract O60). Optical reduced scattering coefficient (μ ') at (μ ') versus optical wavelength (λ) for 10 s exposures at 150, 200 and 975 nm as a function of the measured thermal dose for all sonications 250 W/cm . In Fig. 2, the reduced scattering coefficient (μ ') at 975 nm is plotted as a function of the measured thermal dose for all sonications. Results show that HIFU-induced thermal damage results in changes in scattering at all optical wavelengths from 400-1300 nm (Fig. 1). Furthermore, the reduced optical scattering coefficient in- O61 creases dramatically for exposures exceeding approximately 10^3 cu- In vivo comparison of ultrasound and magnetic resonance mulative equivalent minutes at 43°C (CEM ) (Fig. 2). thermometry for guidance of HIFU mild hyperthermia 2, 1 2 1 2 1 CONCLUSIONS The apparent threshold for optical property changes R. Staruch , S. Sethuraman , B. Cheng , J. Kruecker , R. Chopra in chicken breast tissue is broadly consistent with other studies of Department of Radiology, UT Southwestern Medical Center, Dallas, the thermal dose threshold for lesion formation. AO monitoring of Texas, USA; Ultrasound Imaging & Interventions, Philips Research North HIFU therapy is feasible and this modality may be useful as an alter- America, Cambridge, Massachusetts, USA native to thermometry and dosimetery. Wavelength-dependent op- Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O61 tical property changes can be used to improve the AO sensing of lesion formation during HIFU therapy. [Work supported by the F. V. OBJECTIVES To determine the in vivo feasibility of using strain-based Hunt Postdoctoral Fellowship of the Acoustical Society of America, ultrasound (US) thermometry to monitor mild HIFU heating in muscle the University of Oxford, and EPSRC grant number EP/K02020X/1]. tissue, by direct comparison of temperature measurements made Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 41 of 122 simultaneously using US strain estimation, magnetic resonance (MR) thermometry, and implanted optical sensors. METHODS US thermometry in thigh muscle of anesthetized rab- bits was performed in a 3T MRI (Ingenia, Philips) using a single- element 5 MHz US transducer (Y109, Sonic Concepts) fitted within the central aperture of a single- element 1.1 MHz HIFU transducer (H102, Sonic Concepts); both had a focal distance of 59 mm. A fiber-optic temperature sensor (T1C, Neoptix) was im- planted into the thigh muscle; its location was visualized using 3D T1- weighted MRI (Fig. 1). Transducer location was controlled using an MR-compatible positioning system (RK100, FUS Instru- ments), to set the US focus at offsets of 0, 2, and 4 mm away from the sensor. US pulse-echo acquisition and HIFU energy de- position were interleaved using an open-architecture US system (Vantage 128 HIFU configuration, Verasonics). Filtered US signals were passed into the MR scan room through a grounded RF Fig. 2 (abstract O61). Agreement between MR thermometry and penetration panel. US echo shifts were tracked in the raw RF US fiber optic sensor in image slices along and across the HIFU focus data to derive thermally-induced strains which are proportional to temperature rise. MR temperature maps were calculated using the proton resonance frequency shift technique from RF-spoiled fast field-echo phase images (echo time 12 ms, in-plane reso- lution 2 mm, slice thickness 4 mm) acquired across and along the HIFU focus. The temporal resolutions of the MR, US, and fiber-optic acquisitions were 5, 1, and 1 seconds, respectively. RESULTS Simultaneous US and MR temperature mapping data was successfully acquired and compared with invasive fiber-optic measurements during 15 HIFU sonications in 2 rabbits. MR temperature measurements in a 4 x 4 mm ROI at the location of the fiber optic sensor agreed well with optical measurements, with mean difference and temporal variation less than 0.5°C (Fig. 2). US thermal strain profiles acquired at the location of the HIFU focus 2 to 4 mm away from the sensor correlated well with sensor readings (R = 0.93 ± 0.03, Fig. 3). CONCLUSIONS In in vivo rabbit muscle under normal respiration and perfusion, strain-based ultrasound thermometry is feasible in the mild hyperthermia range. Fig. 3 (abstract O61). Agreement between scaled US thermal strain and fiber optic sensor in rabbit thigh O62 Changes in backscatter of liver tissue due to thermal heating can be used for guiding focused ultrasound ablations 1,2 1,2 V. Barrere , D. Melodelima 1 2 LabTAU, INSERM, Lyon, France; Université Claude Bernard, Lyon 1, Lyon, France Correspondence: V. Barrere Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O62 OBJECTIVES High-intensity focused ultrasound (HIFU) is a noninva- sive therapeutic modality concentrating ultrasonic energy in a small volume of biological tissues. Only in the focal volume the temperature rises above the threshold of thermal coagulation. A non-invasive modality is needed to effectively guide and monitor non-invasive HIFU treatments. Today, magnetic resonance imaging (MRI) and ultrasonic imaging are the two main modalities used in combination with HIFU for guiding the treatment. MRI is superior to ultrasound in visualizing tissue temperature and necrosis but this technique is highly expensive and lacks portability. Ultrasonic im- aging has advantages in its inexpensiveness and portability. In Fig. 1 (abstract O61). 3D T1-weighted MRI of in vivo experiment addition, in most cases the ultrasound imaging probe is placed such setup for simultaneous ultrasound and MR thermometry in rabbits that the imaging plane is aligned with the HIFU acoustic axis. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 42 of 122 Conventional B-mode imaging shows the spatial distribution in amp- METHODS Collection and handling of postsurgical breast specimens litude of the echoes reflected by acoustic impedance mismatches were approved by the institutional review board (IRB) of Columbia and is widely used to guide and monitor HIFU treatments. However, University. Six post-surgical mastectomy breast specimens were ob- B-mode imaging provides limited information about the formation of tained immediately after surgery for HMI imaging. The HMI setup coagulation necrosis due to HIFU. In most cases hyperechoes are vis- consists of a 93-element, 4.5-MHz HIFU transducer confocally aligned ible due to microbubbles generated by either acoustic cavitation or with a 64-element 2.5-MHz phased array to transmit and receive boiling. One of the limitations is that tissue thermal coagulation is through a 4-board VDAS system. The HIFU transducer was driven by not always linked with microbubble generation. In addition, it is diffi- an amplitude-modulated sinusoidal signal to vibrate the tissue at cult to contour precisely the ablated zone based on these hypere- focal area. To generate a 2D/3D HMI displacement map, a point-by- choes. Several methods have been proposed to characterize thermal point raster scan acquisition was used with a step size of 2 mm. At change based on other parameters, such as ultrasonic backscatter. each spot, the focused ultrasound exposure was 0.06 s long (3-cycle One of the oldest methods is the temperature estimation from the oscillations at 50 Hz), during which 60 RF frames at 1-kHz pulse repe- echo shift due to thermally induced change in speed of sound but is tition frequency were acquired for cross-correlation. limited to temperature up to 50-55°C. Thermally induced change in RESULTS 60x15 mm HMI displacement maps could be generated to ultrasonic attenuation has also been studied and used to monitor HIFU map the relative stiffness on the target area within 15 minutes (Fig. treatment, but the change in the attenuation coefficient is due to co- 1) indicating lower displacement in the tumor region and higher dis- agulation and not to the temperature rise. Techniques for estimating placement in the peripheral tissue. In Fig. 1, the average peak-to- the elasticity of tissue are under also investigation based on the fact peak displacement in the tumor defined on the B-mode was found that tissue become stiffer when coagulated but not as a function of the to equal 6.52±3.65 μm, and 38.70±21.79 μm in the surrounding tis- temperature. In this study we investigated the change in ultrasonic sue. A Student’s t-test showed significant difference (P < 0.0001) be- backscattered energy due to the thermal coagulation itself without tween the tumor and peripheral tissue in the HMI displacement. In microbubble generation from 37°C and up to 70°C. the meanwhile, HMI displacement map showed larger relatively stif- METHODS To minimize the cavitation nuclei in the tissue sample, the fer region compared to tumor region on the B-mode. tissue was carefully degassed prior to HIFU exposure. A total of 26 CONCLUSIONS HMI can successfully map the relative stiffness at vari- experiments were performed in porcine liver. Liver samples were able depths on post-surgical human breast mastectomy specimens used because the knowledge of ultrasonic backscatter is largely with or without the skin. The tumor on the HMI appeared slightly larger described. The origin of ultrasonic backscatter from liver tissue is than the one delineated on the B-mode. This study laid the foundation attributed to collagenous septae between liver lobules. Ultrasonic for future clinical study on HMI guided focused ultrasound treatment. backscatter change due to cell death is attributed to the destruction of the cell nuclei. Ultrasonic RF signals at a center frequency of 2.5 MHz backscattered from the tissue before, during and after thermal coagulation due to HIFU exposure at 3MHz were obtained with a pulse-echo transducer and analyzed off-line. The tissue before, during and after the thermal coagulation was also examined by histology using an optical microscope. These results were then com- pared and discussed to clarify the mechanism of the backscattered energy change due to thermal coagulation induced by HIFU. Long exposure time (120 seconds) was used to observe smooth temperature increase from 37 to 70°C. RESULTS The model predicted a linear increase 10 dB. A linear increase 8 dB was measured in ultrasound backscattered power during experiments. The tissue temperature increase estimated using backscattered energy correlated well (r=0.79) with temperature measurements performed using thermocouples. This linear relation- ship between changes in the backscattered energy and actual temperature was observed up to 70°C. CONCLUSIONS Successful temperature estimation may allow creat- ing 2D temperature maps during HIFU treatments. O63 Tumour characterization of human breast mastectomy specimens using Harmonic Motion Imaging (HMI) 1 1 1 1,2 Y. Han , S. Wang , T. Payen , E. Konofagou 1 2 BME, Columbia University, New York, New York, USA; Radiology, Fig. 1 (abstract O63). (a) B-mode image of a post-surgical breast Columbia University, New York, New York, USA mastectomy specimen with tumor appear dark in the yellow dash Correspondence: Y. Han line. (b) HMI displacement overlay on B-mode image with color bar Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O63 showing the HMI displacement. *The HMI scan was not completed due to time constraints according to the IRB protocol OBJECTIVES Breast cancer is the most common cancer as well as the second leading cause of cancer death among women. There is a need to develop a breast imaging technique for reliable identifica- tion and differentiation of breast masses based on stiffness. Recently we have shown that Harmonic Motion Imaging (HMI) can be used to O64 differentiate relative stiffness and monitor HIFU ablations in small Monitoring of nonlinear scattering during cavitation-enhanced lumpectomy human breast specimens. The objective of this study is ultrasonic heating for coagulation detection to apply HMI on post-surgical mastectomy breast specimen with or S. Yoshizawa, K. Tomiyasu, R. Iwasaki, R. Takagi, S. Umemura without skin to mimic the in vivo environment and characterize Tohoku University, Sendai, Japan tumor at different depth for better tumor localization and identifica- Correspondence: S. Yoshizawa tion before and after HIFU treatment. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O64 Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 43 of 122 OBJECTIVES A noninvasive technique to monitor thermal lesion formation is necessary to ensure the accuracy and safety of high- intensity focused ultrasound (HIFU) treatment. Methods for the co- agulation detection and temperature monitoring using ultrasound echo changes, such as the decorrelation of echo signals, have been investigated. On the other hand, methods for the enhancement of HIFU heating using cavitation bubbles have been investigated for the efficient thermal treatment. However, it is difficult to apply the monitoring methods using ultrasound echo changes during cavitation-enhanced HIFU heating because cavitation bubbles tend to produce random echo signals. The objective of this study is to de- Fig. 1 (abstract O64). Schematic of experimental setup velop a noninvasive technique to detect the thermal lesion formation in cavitation- enhanced ultrasonic heating. In this study, nonlinear components of scattered ultrasound from cavitation bubbles are ana- lyzed and used for the coagulation detection. METHODS Figure 1 shows a schematic of the experimental setup. A degassed chicken breast tissue was used as a target tissue. The water was kept at approximately 36°C. HIFU was generated by a 256- element array transducer (Imasonic) with both diameter and focal length of 120 mm. The transducer was connected to 128-ch staircase voltage amplifiers (Microsonic) by electrically combining each two adjacent elements and driven at 1 MHz. A phased array probe (Hitachi Aloka UST-52105) was set in the central hole of the HIFU transducer and connected to a programmable ultrasound imaging system (Verasonics Vantage 256). Figure 2 shows the HIFU sequence consisting of high-intensity short pulses to generate bubble clouds, named “trigger pulses”,and following moderate-intensity long bursts for the enhancement of the ultrasonic heating, named “heating bursts”. The focal point of the trigger pulse was electronically scanned at each corner of a regular hexagon 3 mm each side and a ring focal region was generated employing a sector vortex method in the heating burst exposure to cover the six foci of the trigger pulse for the volumetric cavitation-enhanced heating. The total acoustic powerforthetriggerpulse andheating burst were 1800 and 90 W, re- spectively. The duration and interval time for trigger pulses at each focal point were 25 and 3 μs, respectively. The trigger pulses were laterally Fig. 2 (abstract O64). HIFU sequence consisting of high-intensity short scanned for four times. For heating bursts, the duration and interval time pulses to generate bubble clouds and following moderate- intensity long for trigger pulses at each focal point were 5 ms and 4 μs, respectively. bursts for the enhancement of the ultrasonic heating The focal spot was scanned 5 times. The subtotal durations of trigger pulses and heating bursts were 0.67 and 50 ms, respectively. Immedi- ately after the end of the heating bursts, a 2-ms interval time was re- served for ultrasonic imaging with planewavetransmissionsata frequency of 1.88 MHz. Ultrasonic RFdatawerealsoacquiredduringthe HIFU exposure for the passive coagulation detection. RESULTS Figure 3 shows temporal change in spectral intensity calcu- lated from the RF data during the HIFU exposure after the receive beam- forming with the fixed focus at the HIFU geometric focus. The blue and red lines denotes the acoustic signal intensity at 1 and 2 MHz, respect- ively. Figure 4 shows pulse inversion (PI) images during the HIFU interval time just after HIFU duration started and at the moment when high brightness appeared, 8.0 s after the start of HIFU exposure. The result showed a good correlation between the intensity of the second har- monic HIFU echoes and the emerging high brightness in the PI image. The emerging high brightness was considered as boiling bubbles. It is inferred that the intensity of the second harmonic HIFU echoes in- creased due to newly generated cavitation bubbles which were caused by the trigger pulse reflected by the boiling bubbles. In other words, when the spectral peak was seen, the treatment region was heated suffi- ciently and the cavitation threshold in the surrounding region was de- ceased because of the relatively high temperature. By stopping the HIFU exposure at this moment, overheating would be avoided. CONCLUSIONS In this study, the nonlinear scattering during cavitation- enhanced ultrasonic heating was monitored with an ultrasound im- aging probe. The intensity of the second harmonic HIFU echo increased at the moment when high brightness appeared in the PI image, indi- cating the coagulation in the HIFU focal region. The results will be Fig. 3 (abstract O64). Spectral intensity calculated from the RF data shown and discussed in the presentation when the HIFU exposure is during the HIFU exposure after the receive beamforming with the automatically stopped by detecting the second harmonic intensity fixed focus at the HIFU geometric focus exceeded the certain threshold. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 44 of 122 recorded. The experiment was repeated with various attenuating layers added to the beam path between the target ROI and the FUS transducer. Four different layers with attenuation between 1.2 to 6.3 dB were used. RESULTS The destruction of microbubbles by the treatment beam increases with treatment amplitude and burst length, as expected (Fig. 2). However, any single measurement cannot reliably reveal the in situ beam intensity without controlling the microbubble concentration. The data sets of destruction curves obtained with attenuating layers were matched to the un-attenuated reference data set using a multi- parameter fitting algorithm (Fig. 3). The resulting fitted beam intensity was found to match closely the actual values, verified by independent measurement of attenuation. The errors of in situ beam intensity mea- sured with the proposed method were found to be less than 1.1 dB. CONCLUSIONS Analyzing the complete destruction characteristics produced a feature-rich data set that can be fitted to more than one unknown parameters simultaneously. High frame rate imaging pro- vides crucial speed advantage to this procedure. We proposed a scheme in which, at a pre-treatment phase, a patient is injected with Fig. 4 (abstract O64). PI images during the HIFU interval time just microbubble contrast agent, exposed to low-dose ultrasound from after HIFU duration started and at the moment when high the treatment device, have the destruction characteristics of the bub- brightness appeared bles analyzed to ascertain ultrasound focus and in-situ intensity, and compensation to the therapy planning applied, before the actual course of treatment is applied. This study also demonstrated some O65 capabilities of the in-house designed 2D array therapy system. Par- Estimation and Compensation of in-situ ultrasound intensity using ticularly interesting is that an arbitrary treatment ROI can be exposed a 2D array therapy system and high frame rate imaging in less than 32 ms. In this study the scanning speed of the treatment 1,2 1,2 1,2 1,2 1,2 1,2 Y. Peng ,B.He , N. Deng , X. Chen , S. Chen , C. Chin focus was exploited to ensure that the entire ROI is exposed evenly Biomedical Engineering, Shenzhen University, Shenzhen, Guangdong, in between excessive imaging events. China; National-Regional Key Technology Engineering Laboratory for Medical Ultrasound, Shenzhen, China Correspondence: Y. Peng Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O65 OBJECTIVES We investigated the potentials for ultrasound to provide some patient-specific information which would improve delivery of opti- mal FUS treatment. Ultrasound had been shown to help locate the ac- tual in situ focal point of FUS, and therefore, it is possible to compensate for navigational error due to beam distortion by the heterogeneous hu- man body. However, ultrasound still cannot assist in determining correct ultrasound dosage in a realistic clinical setting. Microbubbles has been investigated as a biocompatible, internal “probe” to convert a local par- ameter to an echo characteristic that can be measured externally. We accessed the main challenge is that the multiple acoustic parameters are not easily isolated from the multiple measureable characteristics of the echo signals (such as frequency shifts and harmonic component). In order to isolate the multiple factors (such as attenuation and perfusion Fig. 1 (abstract O65). The experimental setup rate) contributing to measurable echo characteristics, we sought to ex- ploit the highly specific behaviors of microbubble destruction when ex- posed to intense ultrasound. This paper reports a feasibility study of a pre-treatment scheme to determine effective attenuation and other rele- vant parameters and subsequently compensate for them during the ac- tual therapeutic procedure. METHODS A multi-channel transmission system and an array transducer was designed and built (Fig. 1). The current transmission system provides 128 physical channels that transmit arbitrary waveform with an analog bandwidth of 12 MHz at 4W sustained power or 50W peak power. The number of channels is scalable up to 1024. The transducer consists of a 125-element 2-D array operating at 2.1MHz. The complete system pro- duces a high resolution focal spot of 0.8x0.8x5 mm, steerable to +/- 8 mm in any direction. Highly flexible treatment plans can be implemented with successive focal points can be targeted at an extremely high rate. Thus arbitrary exposure fields can be achieved. A phantom containing Sonovue ® microbubbles are exposed to the treatment beam. The treat- ment focus was scanned to expose an ROI of 8 mm diameter. A programmable high frame rate scanner system (Verasonic Vantage-128, USA) was used to capture echo data at very high spatial and temporal resolutions. A custom imaging sequence produced good quality B-scan frames at a rate of 1 kHz, which are interleaved between FUS exposure at various burst lengths and amplitudes. The treatment beam is normal to Fig. 2 (abstract O65). Destruction curves of microbubble in the the imaging plane. Reconstructed echo image frames were analyzed to treatment ROI for different excitation voltage at a specific locate the treatment ROI automatically. The dynamics of the averaged burst length echo signal in the ROI as functions of treatment parameters was Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 45 of 122 RESULTS Oxy-NGV was developed and characterized to be rod shape with 400 nm length and 100 nm diameter. Through cell-free experi- ment, oxygen inside oxy-NGV can be released by applying 0.8 MPa ultrasound pulse, consequently increasing oxygen concentration in solution immediately. Oxygen releasing rate and amount can be spa- tiotemporally controlled by ultrasound intensity. Cytotoxicity test in MCF-7 and HeLa cell shows increased cell death rate in Oxy-NGV me- diated SDT group compared to pure PpIX one. The improvement of cell death rate is probably attributed to the increased singlet oxygen level as proven by increased SOSG fluorescence intensity. Meanwhile, both the cytotoxicity and singlet oxygen amount have positive cor- relation to the extracellular and intracellular oxygen level by chan- ging Oxy-NGV concentration. In hopxia condition, the therapeutic improvement is more significant than normal oxygen condition. CONCLUSIONS In summary, Oxy-NGV can efficiently deliver oxygen in a presicely controlled manner and consequently enhance SDT out- come through the mechanism of enhancing singlet oxygen produc- tion. NGV, as a novel stable nanometer size contrast agent and oxygen carrier, has great potential to enhance other oxygen medi- ated cancer therapy like radiotherapy, chemotherapy and photo- dynamic therapy. Fig. 3 (abstract O65). Extent of destruction measures at various incident intensities and burst lengths. Note that both axes are not linearly scaled O67 Synergistic ablation of tumours in vivo by high-intensity focused ultrasound and ethanol H. Murad, G. Halliburton, D. Luo, H. Yu, D. Khismatullin O66 Biomedical Engineering, Tulane University, New Orleans, Louisiana, USA Enhanced sonodynamic therapy using oxygen-rich nano gas Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O67 vesicle Y. Yang, X. Hou, L. Sun OBJECTIVES High-intensity focused ultrasound (HIFU) emerges as a Interdisciplinary Division of Biomedical Engineering, Hong Kong powerful technology for noninvasive or minimally invasive non- Polytechnic University, Hong Kong, China ionizing treatment of cancer, with recent FDA approval. HIFU de- Correspondence: Y. Yang posits a large amount of acoustic energy at the focal region within Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O66 the target tissue (i.e., tumor), causing tissue heating and necrosis, a process known as thermal ablation. Noninvasiveness is an important OBJECTIVES Sonodynamic therapy (SDT), based on the synergistic ef- advantage of HIFU over other thermal ablation methods, and our la- fect of low intensity ultrasound and sonosensitizer, is a potential boratory explores the ways for synergistic combination of HIFU with noninvasive approach for the treatment of cancers. Singlet oxygen, other therapeutic modalities to achieve the complete destruction of the major cytotoxic agent, is generated from dissolved oxygen within large and multifocal tumors. In this study we test the hypothesis that treatment region. However, because of the hypoxia microenviron- HIFU and percutaneous ethanol injection (PEI), a leading method for ment of solid tumor, oxygen deficiency restricts the singlet oxygen chemical ablation, have a synergistic effect on ablation of aggressive generation and consequently dramatically decrease therapeutic ef- liver and prostate cancers in vivo. fect. To locally increase the oxygen level may potentially improve the METHODS This in vivo study was performed using the xenograft treatment outcome. Here, we reported an oxygen enhanced SDT mouse models of human liver and prostate cancers. Hep3B human method utilizing oxygen-rich nano gas vesicle (Oxy-NGV). The objec- cancer cells and DU145 human prostate cancer cells (2.0×106) were tives of this study is firstly to develop oxygen-rich nano gas vesicle injected on flanks of athymic nude mice. Tumors were allowed to and evaluate the oxygen releasing mechanism, then to investigate grow to 8-10 mm size and then separated into the following treat- the improvement of the new oxygen-rich nano gas vesicle enhanced ment groups: HIFU alone, PEI (50%Etoh, 50 μl) alone, PEI+HIFU sonodynamic therapy as well as the underlying mechanism. If this (50%Etoh, 50 μl), and sham. Tumor sizes were measured by caliper Oxy-NGV based oxygen delivery strategy proven to be efficient, it will every day and a veterinary diagnostic ultrasound system was used have great potential to extend applications to other oxygen based pre-treatment, 5 days, and 12 days’ post- treatment. Tumor volumes cancer therapy like radiotherapy, chemotherapy and photodynamic were calculated from the ellipsoid formula V=πabc/6, where a, b, c therapy. are tumor sizes in three orthogonal directions. Tumors were surgi- METHODS Oxy-NGV was developed based on the nano gas vesicle cally removed and fixed using 10% formaldehyde solution. Samples (NGV) produced by cyanobacteria. It was then characterized regard- were sent for H&E staining with a single blinded pathologist, and ing basic nanoparticle’s property and more importantly the oxygen live/dead percentages of tumor cross sections were determined at 5 releasing efficiency through being broken by ultrasound pulse (0.8 and 12 days post treatment. Cryogenic-Scanning Electron Microscopy MPa). The therapeutic effect was evaluated by in vitro cytotoxicity by (Cryo-SEM) was also used to capture membrane disruption post HIFU ultrasound treatment for 5 mins with the intensity of 5W/cm after +PEI exposure on DU145 prostate cancer cells. MCF-7 and HeLa tumor cells were incubated with sonosensitizer Pro- RESULTS Tumor growth is significantly reduced or completely elimi- toporphyrin IX (PpIX) and Oxy-NGV. Then the singlet oxygen level, as nated in tumors treated with HIFU in combination with PEI (Fig. 1). the major cytotoxic agent, was imaged using Singlet Oxygen Sensor Tumors treated with HIFU alone showed a decrease in tumor size at 5 Green (SOSG) in both cell-free model and intracellular scenario. days, then rebounding to similar sizes as the sham. PEI alone tumors Meanwhile the oxygen level was also tested by dissolved oxygen showed no significant reduction in size and continued to grow. Hist- meter compared with conventional SDT method. These studies were ology shows largest necrotic tissue area in tumors treated with PEI and repeated in both normal oxygen level and hypoxia condition. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 46 of 122 HIFU at 5 and 12 days’ post treatment. Cryo-SEM images show large was delivered via either H105, an unfocused 52 mm disc transducer, macropore (>1-2 micrometer) formation in cells treated with PEI+HIFU. or H185D, a 49 mm disc transducer with three cylindrical lenses, each CONCLUSIONS The combination of HIFU and PEI shows a synergistic focused to a depth of 20 mm from the exit plane. Our US pulse dura- effect on tumor destruction at very low concentration of ethanol and tions spanned 19 μs–22 ms, with PNPs spanning 0.6–6.9 MPa. 24 lower acoustic power. This combination may become an effective hours after surgery, pigs were sacrificed to harvest treated and con- minimally invasive treatment option that’s safer for patients with liver trol liver lobes. After sectioning, spatially- mapped samples were ana- or prostate cancer. Utilizing these two FDA approved treatment in lyzed for luciferase expression. combination could lead to a disruptive and translational method of RESULTS Our ongoing experiments have added further support for a tumor treatment. species-generalized model that increasing pulse duration enables lower PNP for effective UMGD. Within a paired study, increasing pulse duration used with H185D from 19 μs to 200 μs at a constant 6.9 MPa PNP yielded up to 17-fold increases in sampled luciferase ex- pression. In a repeated H185D study, we have also shown an increase in expression using lower-pressures and longer pulse durations (200 μs, 4.5 MPa and 2 ms, 2.7 MPa groups vs. the 19 μs, 6.9 MPa group). Nevertheless, pulse durations above 2 ms paired with lower pressure did not appear to further enhance expression. Despite these expres- sion increases, ALT and AST values were comparable or lower in groups using longer pulse durations compared with groups using a 19 μs pulse duration. Experiments using H105 yielded a similar trend. Relative to an 18 μs, 2.7 MPa condition, we found increased expres- sion in groups with conditions of 1 ms, 1.2 MPa; 4 ms, 0.8 MPa; as Fig. 1 (abstract O67). LEFT: Prostate tumor (8x9mm) before PEI well as 22 ms, 0.5 MPa. Our 4 ms, 0.8 MPa group resulted in elevated +HIFU ablation, RIGHT: Tumor was eliminated at 2 weeks with AST levels, however changes in ALT and AST values of all other groups no re-occurrence were nominal. From these experiments, we analyzed several spatial considerations between the two transducer designs. Of the two, the unfocused H105 treats a larger tissue volume simultaneously, but in ex- change it generates lower PNPs at maximum operational power. Con- versely, the cylindrically-focused H185D treats a condensed tissue volume which allows higher peak pressures at its operational limit. H105 outperformed H185D when comparing conditions of equivalent pulse duration and PNP, a reasonable result given H105’slargervolume. When instead comparing the two transducers in terms of average in- tensity across their entire active area, we found no obvious discrepancy in expression. However, when using H185D in its higher range of PNPs (4.5-6.9), individual tissue segments were observed to have higher max- imum expression than the maximum values sampled using H105 at its limit PNP of 2.7 MPa. We are currently investigating whether this stems Fig. 2 (abstract O67). LEFT: Liver tumor (7x10mm) before PEI from the higher maximum PNP generated by H185D or whether using +HIFU ablation, RIGHT: Tumor was eliminated at 2 weeks with the transducer at those higher powers recruits greater UMGD efficacy no reoccurrence from weakly-focused areas. CONCLUSIONS By manipulating US pulse durations, our group achieved increased gene expression following UMGD in pigs. Such O68 tuning has also allowed comparable expression at decreased PNPs, Multivariate ultrasound signal manipulation for effective gene circumventing voltage limitations of piezo-materials. Since attenu- delivery to pig livers ation impedes high PNPs in transcutaneous UMGD applications, 1 1 1 1 2 J. Harrang , S. Song , M. Kajimoto , J. Chen , R. Fu1, K. Morrison , these results have promising implications for that modality. Our re- 2 1,3 G. W. Keilman , C. H. Miao sults demonstrate the advancement of efficient gene transfer in large Center for Immunity and Immunotherapies, Seattle Children's Research, animal models and we have also begun to elucidate the spatial re- Seattle, Washington, USA; Sonic Concepts Inc., Bothell, Washington, quirements for effective UMGD. USA; Pediatrics, University of Washington, Seattle, Washington, USA Correspondence: J. Harrang Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O68 OBJECTIVES Over several years our group has increasingly shown O69 that ultrasound-mediated gene delivery (UMGD) can be enhanced by Destruction of staphylococcus aureus biofilms on surgical mesh selecting favorable ultrasound (US) parameters. In particular, increas- T. A. Bigelow, H. Wu, C. Thomas ing pulse duration lowers the peak negative pressure (PNP) required Iowa State University, Ames, Iowa, USA for UMGD in mouse and cell models. This effect allows selection of Correspondence: T. A. Bigelow conditions with minimal associated tissue damage and enables tun- Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O69 ing to maximize the capabilities of piezo-materials. We have now fur- ther investigated scalability of this finding in pigs. In parallel, we OBJECTIVES It may be possible to significantly reduce the ~35,000 examined several spatial effects which represent important consider- surgeries each year needed to replace infected surgical meshes fol- ations for eventual clinical application of this novel technology. lowing abdominal hernia repair. The use of a surgical mesh has be- METHODS We have established an open surgery protocol used to ex- come standard medical practice. However, mesh infection is a plore optimal US parameters for efficient UMGD in pigs. First, the significant complication with incidence rates ranging from 1% to liver of each pig was exposed via a midline incision. Next, using con- over 10%. Mesh infections require reoperation for mesh removal trast US to confirm placement and perfusion, we catheterized a spe- ~70% of the time resulting in the potential for hernia reoccurrence cific branch of the portal vein. Just prior to therapeutic US exposure, and the need for additional operations. Therefore, there is a critical the inferior vena cava was temporarily occluded. Then US exposure need to develop new methods to noninvasively treat mesh infections and infusion of a solution containing pGL4 plasmid and phospholipid without removing the mesh. Our goal is to develop ultrasound microbubbles (MBs) were initiated simultaneously. Therapeutic US cavitation-based histotripsy to treat infections on surgical mesh. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 47 of 122 METHODS S. aureus biofilms were grown on 1 cm square surgical mesh samples for 3 days. The samples were then rinsed with phos- phate buffered saline prior to being inserted into Aquaflex Ultrasound Gel Pad Standoffs (Parker Laboratories Inc., Fairfield, NJ). The gel pads are bacteriostatic to minimize bacteria growth and have approximately the same mechanical properties as abdominal muscle. Mechanical properties have a significant impact on cavitation treatment (such as those used in our study), and therefore it is important they be relatively similar to real biological tissue. Each gel pad was cut in half allowing for two experiments per gel pad, and a slit was then cut in the gel pad to allow the insertion of the infected mesh sample. The focus of a spherically focused transducer (1.1 MHz, 12.9 cm focal length, 12.7 cm diameter) was then aligned on the mesh samples using a low-power signal from a pulser-reciever (Panametrics 5900, Olympus Corporation, Tokyo, Japan). Once aligned, the mesh samples were exposed to either a sham expsosure or histotripsy pulses (compressional pressure of 155 MPa, rarefactional pressure of 17 MPa) with tone burst durations of 3, 5, or 10 cycles at a pulse repetition frequency of 333 Hz for a duration of 15 seconds per exposure location with 5 repetitions per exposure group including the sham. The entire mesh was treated by scanning the focal spot in a raster pattern over the mesh using a step size of 750 Fig. 2 (abract O69). The number of S. aureus CFUs left on the gel μm. After treatment, the number of colony forming units (CFUs) on the pad following the ultrasound exposures mesh and the surrounding gel was independently determined. RESULTS For the mesh samples (Fig. 1), sham exposures have statisti- cally significantly more colony forming units than each of the treatment O70 groups. The absence of a bar corresponds to when no CFUs were A special retinal ganglion cell responses to low-frequency focused found. If we compare the means, we see a reduction of 2.00-log10 for ultrasound stimulation the 3-cycle treatment, 3.25-log10 reduction for the 5-cycle treatment, H. Zhao, Q. Jiang, G. Li, M. Su, H. Zheng, W. Qiu and a 3.23-log10 reduction for the 10-cycle treatment. However, the Shenzhen Institutes of Advanced Technology, Chinese Academy of differences between the 3-cycle treatment and 5- cycle/10-cycle treat- Sciences, Shenzhen, China ments are not statistically significant once a Bonferroni correction has Correspondence: H. Zhao been applied. More observations are needed to determine if the 3- Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O70 cycle exposures are significantly higher. The results from the gel pad (Fig. 2) showed no significant change in the number of CFUs for re- OBJECTIVES Acoustic retinal prosthesis has been put forward using leased bacteria from the biofilm for any of the treatments. However, high-frequency US with noninvasive and high- resolution advantages. the high variance in the numbers of CFUs for the gel samples means But its application is limited by fabrication, energy consumption and that while these numbers are not significantly different, we cannot mismatching to human eyeball’s anteroposterior axis. As our previous claim that additional bacteria are not being released by the ultrasound study demonstrated, the spatial resolution of low-frequency focused exposures. Given that approximately 40% of the bacteria in the sham US (LFUS) can be improved by decreasing applied acoustic intensity. experiments were on the mesh, we would need to show that the bac- Thus we prefer the acoustic retinal prosthesis using LFUS and are in- teria in the gel are not statistically greater than this to make this claim. terested in the electrophysiological properties of retinal ganglion cell This would require more observations per treatment group. (RGC) responses. This study has inspected the characteristics of one CONCLUSIONS The ultrasound histotripsy treatments are effectively special type of RGCs’ responses to LFUS in comparison with their destroying most of the biofilm on the infected surgical mesh. We ex- light responses, and examined the response changes in the presence pect that further optimization of the exposure parameters will further of ON pathway blocker. enhance destruction of the biofilm. METHODS A 2.25 MHz focused US transducer (D=0.75 in., SF=2.0 in.) was used to stimulate retina which was cultured in a multi-electrode array system (MEA2100, MCS, Fig. 1a). The acoustic property was evaluated by hydrophone (UMS3, Precision acoustics). US stimulation was modulated at pulsed mode (Fig. 1b). Light stimulation was mod- ulated in the same mode to give a uniform field flashes. The electro- physiological data collected from MEA was detected for neural spikes and sorted by Plexon Offline Sorter. Only channels recording single- cell activities were adopted for subsequent analysis. Peri-stimulus time histograms and raster plots were plotted for each RGC using Spike 2. RESULTS LFUS can activate RGC and generate sustained excitation (Fig. 1c). A comparison between light and US- induced responses in the same RGC shows differences in response temporal pattern and polarity (Fig. 1d). Light produces sustained excitation at stimulus onset and prolonged inhibition at offset, which means the recorded RGC is ON-centered. But US elicits a transient peak of excitation followed by delayed sustained excitation at stimulus onset, and transient excitation at offset. 21 investigated RGCs show the same differences. A typical result of ON pathway block- ing shows the delayed sustained excitation is eliminated by 100 μM L-AP4 (Fig. 1e). It is possible that this sustained excitation is generated from photoreceptors which transmission to bipolar Fig. 1 (abract O69). The number of S. aureus CFUs left on the mesh cells is blocked. The exemption of ON- and OFF-transient excita- following the ultrasound exposures. The absence of a bar indicates tion are probably because US directly activates interneurons or no CFUs remained on the mesh RGCs in retinal neural circuits. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 48 of 122 CONCLUSIONS We have investigated the neurophysiological properties of RGC responses to LFUS. We have discovered some new temporal re- sponse patterns of RGCs that haven’t been reported previously, including the characteristic dual-peak response patterns in ON-sustained RGCs and transient RGCs. In addition, we have found that US can modulate the temporal-spatial characteristics of RGC firing activities, which suggests that US can encode the information transmitted by RGCs. These results of our study will provide an important foun- dation for the development of ARP. Fig. 1 (abstract O71). See text for description Fig. 1 (abstract O70). See text for description O71 Guided longer pulses from a diagnostic ultrasound and intraclot microbubbles enhanced catheter directed thrombolysis 1 1 1 1 1 2 S. Gao , Q. Zhu , X. Dong , Z. Chen , Z. Liu , F. Xie Department of Ultrasound, Xinqiao Hospital, Third Military Medical University, Chongqing, China; Internal Medicine Fig. 2 (abstract O71). See text for description Cardiology, University of Nebraska Medical Center, Omaha, Nebraska, USA Correspondence: S. Gao O72 Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O71 Effects of f-number on the histotripsy intrinsic threshold and cavitation bubble cloud behaviour OBJECTIVES Insufficiency of MB in and around the vessel- E. Vlaisavljevich, T. Gerhardson, T. Hall, Z. Xu obstructing thrombi significantly reduces the effectiveness of University of Michigan, Ann Arbor, Michigan, USA ultrasound assisted thrombolysis (UT). Combined with intraclot in- Correspondence: E. Vlaisavljevich fusion of MB, guided longer pulses ultrasound from a diagnostic Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O72 transducer should be able to improve catheter directed thromb- olysis (CDT) procedure. OBJECTIVES Histotripsy is an ultrasonic ablation method that frac- METHODS In a thrombo-embolised rabbit IVC model, parallel to tionates soft tissue through the precise control of acoustic cavitation. catheter directed rt-PA thrombolysis procedure, guided moderate MI Previous work has demonstrated that a cavitation cloud can be longer pulses from a modified diagnostic ultrasound transducer formed by a single acoustic pulse with one high amplitude negative combined with an intraclot infusion of MB were aplied to facilitate cycle when the negative pressure exceeds an intrinsic threshold of CDT. The thrombolysis efficacy score, pre and post-treatment plasma ~25-30 MPa. Although previous work has provided significant insight concentration level of D-Dimer, a product of fibrinolysis, were into the process of intrinsic threshold histotripsy, the majority of acquired and compared in the four groups (CDT+UT, CDT alone, UT these studies have used highly focused (i.e. f-number<0.6) transduc- alone, & control). ers. In this study, we investigate the effects of f- number on the his- RESULTS The higher thrombolysis efficacy score (Fig. 1) and consist- totripsy intrinsic threshold and cavitation bubble cloud behavior, ent elevated post-treatment plasma concentration level of D-Dimer which is essential to the development of histotripsy for different clin- (Fig. 2), both indicated a superior of CDT + UT over CDT/UT alone. ical applications. There were no evidences of thrombo-embolism or local thrombus METHODS The effects of f-number on the histotripsy intrinsic thresh- formation in the cardiac-pulmonary vessels. old and cavitation bubble cloud behavior were investigated using a CONCLUSIONS Combined with intraclot infusion of MB, guided lon- 235-element 500 kHz array transducer, with the effective f-number of ger pulses from a diagnostic transducer was able to improve catheter the transducer varied from 0.51 to 0.89 by changing the active ele- directed thrombolysis procedure. This strategy has a possibility to ments in the array. Ultrasound pulses of 1-2 acoustic cycles were ap- achieve earlier clot removal, lower dosage of thrombolytic agent ad- plied to tissue mimicking phantoms, and the resulting cavitation ministrated, and in consequence lower incidence of thrombolysis re- activity was detected and characterized by passive cavitation detec- lated side effects. tion and high-speed photography (Phantom V210, Vision Research). Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 49 of 122 The intrinsic threshold at each f-number was defined as the pressure generated atdifferent microbubble-cell distances remained to be eluci- at which the probability of generating cavitation was 0.5. Optical im- dated. The goal of this study was to investigate how the microbubble- ages were further analyzed to determine the effect f-number on bub- cell distances influence the cell dynamics, such as the cytoskeleton ar- ble cloud characteristics including the bubble cloud dimensions, the rangement and the cellular membrane permeability, and the experimen- “bubble density” within the cloud, and individual bubble size. Finally, tal observations were compared with theoretical simulations. the effect of f-number on histotripsy fractionation efficiency was in- METHODS The in situ cellular responses (e.g., cytoskeleton arrange- vestigated by applying histotripsy to tissue phantoms embedded ment and intracellular delivery) to microbubble-mediated sonopora- with a layer of red blood cells, with the resulting fractionation visual- tion process generated withdifferent microbubble-cell distances were ized using optical imaging. systemically assessed based on an integrated system combining ultrasound exposure apparatus with real-time fluorescencemicro- RESULTS The intrinsic threshold did not significantly change with f- scope imaging. The microstreaming and shear stress generated by number, with the threshold remaining ~27-30 MPa for all conditions an oscillating microbubble was simulated based on an encapsulated (Fig. 1A). The predictability of intrinsic threshold histotripsy was fur- microbubble dynamic modelwith considering nonlinear rheological ther demonstrated by experiments showing close agreement be- effects of both shell elasticity and viscosity. tween the predicted and experimentally measured bubble cloud RESULTS The results show that: (1) as the microbubbles get closer to dimensions for all f-numbers. Quantifying the size of individual bub- cells, the disassembly of the cytoskeleton will accelerate; (2) as the bles formed directly above the intrinsic threshold at different f- num- microbubbles get closer tocells, the permeability of cell membrane bers showed no significant change in bubble size (~300 μm) with f- will have significant improvement; (3) the maximum microstreaming- number. Comparing bubble clouds at different f-numbers showed a induced shear stress on the cell membrane increasesrapidly with re- significant reduction in the “bubble density” with increasing f- ducing the bubble-cell distance. number, ranging from 39.6±3.8 bubbles/mm for an f-number of CONCLUSIONS In summary, by performing live cell fluorescent im- 0.51 to 1.5±0.3 bubbles/mm for an f-number of 0.89 (Fig. 1B). Finally, aging over the process of sonoporation, the accelerating cytoskel- experiments comparing the efficiency of histotripsy ablation demon- eton disassembly and improvedmembrane permeabilization could strated a significant increase in treatment efficiency for lower f- num- result from microbubble-mediated sonoporation with reducing bub- bers. For example, the number of pulses required to fractionate 50% ble-cell distance. The shear stresses resulting from microstreaming- of the treatment zone increased from 14.5±9.5 pulses for an f- generated nearby pulsating bubbles with different bubble-cell number of 0.51 to 209.3±17.4 pulses for an f-number of 0.89, corre- distances were simulated theoretically, which could provide a better sponding to a >14-fold increase in treatment efficiency (Fig. 1C). explanation for observed phenomena. The result suggests that in CONCLUSIONS The results of this study demonstrate that the order to achieve more efficient sonoporation effect in therapeutic ap- histotripsy intrinsic threshold does not significantly change with f- plications, it is better to find and optimal bubble-cell distance number. In addition, results show that histotripsy fractionation andmanipulate the microbubble location more rationally under cer- efficiency decreases at higher f-numbers due to a decrease in the tain conditions. “bubble density” within the bubble cloud. Overall, this study provides significant insight into the effects of f-number on intrinsic threshold histotripsy that will help to guide the design of histotripsy transduc- ers for specific clinical applications. O74 A pilot study of histotripsy using 1.1/2.2 MHz dual-frequency superimposition focused ultrasound pulses Y. Li, R. Wang, M. Lu, W. Huang, F. Ma, L. Zhang, M. Wan The Key Laboratory of Biomedical Information Engineering of Ministry of Education, Department of Biomedical Engineering, School of Life Science and Technology, Xi’an Jiaotong University, Xi’an, China Correspondence: Y. Li Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O74 OBJECTIVES Kidney cancer is a severe disease which can be treated using a non-invasive, controllable and focused ultrasound surgery method, termed as histotripsy. However, the time of le- sion formation following a single frequency histotripsy is long, so the dual-frequency mode histotripsy using second-harmonic superimposition hundred- microsecond pulses to reduce the le- sion time is very necessary. The aim of this research is to explore the feasibility of enhancing histotripsy by increasing effective Fig. 1 (abstract O72). See text for description cavitation and efficient boiling. METHODS By controlling the ratio of dual-frequency acoustic powers, the superimposition of two frequency pressures results in 9 split foci O73 along beam axial within confocal region, and the maximal peak in- Sonoporation-induced intracellular delivery and cytoskeleton tensity of split focus can reach about 2 times the sum of two fre- disassembly with different microbubble-cell distances quency intensities, indicating strong wave interference. Meanwhile, Maochen Wang, Chenliang Cai, Pengfei Fan, Dongxin Yang, Zhiyang Jin, the cavitation threshold lowers and the nonlinear effect strengthens Juan Tu, Xiasheng Guo, Dong Zhang by the dual-frequency superimposition mode. The efficient boiling School of Physics, Nanjing University, Nanjing, Jiangsu, China mechanism becomes dominant in histotripsy. The experiments im- Correspondence: Maochen Wang plemented in polyacrylamide phantoms with bovine serum albumin Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O73 and in porcine kidney ex vivo. The simplified two-stage hundred- microsecond pulsing scheme was used. The lesion formation process OBJECTIVES Sonoporation mediated by ultrasound-driven microbub- in the BSA phantom was monitored by high-speed photography and bles is being extensively studied as a promising technology to facilitate passive cavitation detection. gene/drug delivery to cells, and microstreaming generated by pulsating RESULTS The lesion inception time in gel-phantom was about 0.3s, microbubble near the cell membrane is regarded as one of the most im- which was 6 times shorter than that in single frequency mode (Fig. 1a). portant mechanisms in the sonoporation process. Thepresence of micro- The enhanced pressure, the lowered cavitation threshold and the bubbles is believed to help enhance ultrasound-induced bioeffects, but strengthened nonlinear effect by dual-frequency superimposition re- an in-depth understanding of cellular responses to sonoporation sulted in the lesion inception time decreased and boiling bubbles Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 50 of 122 emerged frequently in the axial split foci during each lesion formation, indicating enhanced histotripsy (Fig. 1b). The interaction of boiling bub- bles between split foci also had contributed to the lesion formation and the lesion size increased both radial and axial directions. The final lesion exhibited a long tear shape with smooth border (Fig. 1c). The le- sion generated in ex vivo porcine kidney was shown in Fig. 2, and the voids appeared with no marked thermally coagulated component remaining after the homogenate had been removed. The root mean square (RMS) amplitude of the broadband noise from filtered passive cavitation detection (PCD) data revealed the strong inertial-cavitation activities, and the increase of RMS demonstrated that the boiling bub- bles arose (Fig. 3a). Meanwhile, the inertial cavitation effect transferred to a higher band increment, which was beneficial to the frequency ab- sorption efficiency (Fig. 3b). CONCLUSIONS This study demonstrated the feasibility of enhancing Fig. 2 (abstract O74). Representative gross morphology of the dual- histotripsy by increasing acoustic intensity, lowering cavitation thresh- frequency-superimposition histotripsy lesions induced in ex vivo porcine old and strengthening nonlinear effect using dual-frequency superim- kidney with dimensions of approximately 6.0×2.8 mm (axial×lateral) position focused ultrasound pulses. The increase of effective cavitation and boiling dominated in the lesion formation. The split foci occur by controlling the ratio of dual-frequency acoustic powers. The maximal peak intensity of split focus can reach about 2 times the sum of two frequency intensities, indicating strong wave interference. The lesion in- ception time decreased, compared with single frequency mode. The final lesion exhibited a long tear shape with smooth border. The root mean square amplitude of the filtered passive cavitation detection data revealed the strong inertial-cavitation activities, and the increase of RMS demonstrated the boiling bubbles arose. Fig. 3 (abstract O74). (a) RMS amplitude of the PCD signals using dual frequencies of 1.1/2.2 MHz on the phantom at different times. (b)Broadband noise in the frequency domain using dual frequencies of 1.1/2.2 MHz on the phantom O75 Ultrasound-guided high intensity focused ultrasound ablation for diffuse adenomyosis J. Chen, Y. Feng, W. Chen College of Biomedical Engineering, Chongqing Medical University, Chongqing, China Correspondence: J. Chen Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O75 This abstract is not included as it has already been published: Feng Y, Hu L, Chen W, Zhang R, Wang X, Chen J. Safety of ultrasound-guided high-intensity focused ultrasound ablation for dif- fuse adenomyosis: A retrospective cohort study. Ultrason Sonochem. 2017; 36: 139-145. Available from: http://www.sciencedirect.com/sci- ence/article/pii/S1350417716304096. O76 Preliminary results of synthetic aperture imaging using random phased array M. Zubair, R. J. Dickinson Bioengineering, Imperial College London, London, United Kingdom Correspondence: M. Zubair Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O76 Fig. 1 (abstract O74). (a) The lesion inception time in gel-phantom OBJECTIVES Randomized phased arrays have been used for generat- was about 0.3s. (b) Boiling bubbles with about 1mm diameter in ing and steering single focus and multiple foci with low levels of phantom acquired by high-speed photography. (c) The final lesion grating lobes due to the breakage of periodicity of the elements and exhibited a long tear shape with dimensions of approximately are considered as useful source of HIFU. However, the reliance of 8.2×1.6 mm (axial×lateral) HIFU on MRI for real time visualization of the targeted tissue is a Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 51 of 122 major constraint in its clinical use due to the high cost of MRI and its low temporal resolution. There is a need to study the imaging cap- abilities of a therapeutic random phased array transducer for guiding the treatment process. METHODS Dual mode ultrasound phased arrays would have the ad- vantage of using the same array for both therapy and imaging due to the inherent registration between imaging and therapeutic frames of reference. The random spherical array would have limited field of view due to the fact that the array is optimized for therapy only and has large, directive elements sparsely positioned on a spherical sur- face. Nevertheless, images obtained will be useful for directing ther- apy as they will be perfectly aligned with the therapy transducer. Since strong scattering objects in path of HIFU beam are also in path of imaging beam, such scattering objects can be detected in real time and the HIFU beam can be adjusted accordingly. In our HIFU system the elements are randomly distributed with inter-element spacing much more than the required half a wavelength for reduced side lobes (Hand et. al. 2009), thus the spatial resolution of this sys- tem is poor. However, we use synthetic aperture imaging technique which has the potential to improve the spatial resolution of the ran- Fig. 2 (abstract O76). Grayscale images of wire target array using dom phased array. The simulations were carried out in MATLAB. The STA imaging approach numerical results were performed for a 1 MHz 256-element random phased array, made by Acublate Ltd, London, UK. Simultaneous foci were generated in simulations as well as experimentally based on the theory described by Gavrilov and Hand (2000a) (Fig. 1). RESULTS For imaging, preliminary simulations of synthetic aperture imaging with a 1MHz 256 element random phased array are shown. In Fig. 2, grey scale image of a wire target array is shown, which is a composite of point spread functions (psfs) at different on- and off- axis positions. The axial spacing between two wires is 10 mm, whereas the lateral distance is 5mm (figure 30. The -6 dB full width half maximum of the focused psf is 1.6 mm. Sub-apertures are being used to image the field of interest, where each sub-aperture contains only those elements which contribute to the pixel being imaged. This not only improves the resolution by decreasing the side lobes level but also reduces the computation time. CONCLUSIONS It was observed that random phased arrays are cap- able of therapy as well as imaging for treatment guidance. The use of sub-apertures improves the resolution and reduce the computa- tion time, however, it also limits the imaging field of view Fig. 3 (abstract O76). Lateral cross section at a depth 130 mm O77 Image-guided blood-brain barrier opening with focused ultrasound and microbubbles in mice using passive microbubble imaging 1 1 1,2 M. T. Burgess , I. Apostolakis , E. Konofagou Biomedical Engineering, Columbia University, New York, New York, USA; Radiology, Columbia University, New York, New York, USA Correspondence: M. T. Burgess Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O77 OBJECTIVES Blood-brain barrier (BBB) opening with focused ultrasound (FUS) and microbubbles is a technique for targeted drug delivery to the brain and has led to the development of ultrasound-guided focused ultrasound (USgFUS) systems for appropriate treatment monitoring and guidance. Comprehensive detection of FUS-stimulated microbubble ac- Fig. 1 (abstract O76). Distribution of 4 simultaneous foci at a depth tivity is critical for successful implementation of these systems. Current of 130 mm techniques passively image microbubble- related acoustic emissions Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 52 of 122 with ultrasound arrays to spatially map the magnitude and location of OBJECTIVES Lipid-coated contrast agents have attracted much atten- microbubble activity throughout the focal volume of the FUS. In turn, tion for contrast ultrasound molecular imaging as well as targeted this provides information about the extent and location of BBB open- treatment, and the connection between enhanced contrast capability ing. Current passive cavitation imaging methods with linear arrays suf- and acoustic properties of the lipid-coated ultrasound contrast fer from poor axial image resolution due to the use of long FUS pulses agents need to be elucidated. and asynchronous transmit and receive sequences that degrades abso- METHODS In this study, microbubbles were fabricated using thin- lute time-of-flight information. The objective of this study was to pre- film hydration and mechanical agitation. Then, the morphology and serve absolute time-of-flight information by using short pulses of FUS, distribution of these microbubbles were investigated through trans- along with synchronous transmit and receive acquisition, for utilization mission electron microscopy (TEM) imaging and dynamic light scat- of image reconstruction similar to pulse- echo B-mode imaging. It is ex- tering (DLS) sizing technology. To demonstrate physical properties of pected that this new passive microbubble imaging (PMI) technique will microbubbles, inertial cavitation threshold and acoustic attenuation improve image resolution and offer a methodology capable of high measurements were carefully assessed and shell parameters were resolution mapping of BBB opening. further estimated. The imaging function of synthesized microbubbles METHODS PMI was carried out during BBB opening with FUS and was also compared with that of SonoVue microbubbles in vivo. microbubbles in a mouse model. An 18-MHz imaging array (L22-14v RESULTS The results showed that the synthesized microbubbles had LF, Verasonics, Inc.) and 1-MHz FUS transducer were submerged in a a spherical shape, a smooth, consistent membrane and uniform dis- tank of degassed water and placed approximately 2 cm above the tribution, with average diameter of 1.484 μm. Imaging studies mouse skull for transcranial application of FUS. The FUS transducer showed that while exhibiting comparable imaging ability, synthe- was aligned at an acute angle relative to the imaging array’s axis. A sized microbubbles had a longer circulation time and better stability research-based ultrasound system (Vantage 256, Verasonics, Inc.) was than SonoVue. Physical characterization showed that compared with used to operate the imaging array in a passive mode and SonoVue, synthesized microbubbles with smaller interfacial tension and synchronize the FUS transmission with receive acquisition. Short dilatational viscosity, indicating less attenuation during the propagation pulses of FUS (2-3 cycles, 200-400 kPa) at a pulse rate of 500-5000 Hz of sound wave. In comparison with SonoVue, IC threshold of the micro- were used to insonify intravenously injected microbubbles as they bubbles are prominently higher in both concentration ranges, explain- flowed through the mouse brain microvasculature. In this scenario, ing better stability of the microbubbles. Present study built a bridge the FUS pulses were used for the dual purpose of imaging and ther- between physical properties and in vivo imaging performance of syn- apy (BBB opening). Receive acquisition frames were recorded for thesized microbubbles, which could provide guidance to the design each FUS transmit and the data was saved for off-line processing. and safe application of ultrasound contrast agents. PMI images were formed using a custom GPU-based image recon- CONCLUSIONS While the average grey scale of various organs all in- struction algorithm in MATLAB (The Mathworks, Inc.). The time delays creased following microbubbles application, tumor imaging showed for implementation of delay-and-sum beamforming were calculated that synthesized microbubbles stayed in the tumor area for longer to account for the propagation from the FUS transducer, to the period of time and has a longer enhancing time. The long-circulating mouse brain, and back to the imaging array. Blocks of PMI frames behavior showed that synthesized microbubbles may be better suited were further processed using spatiotemporal filtering to isolate for tumor imaging and therapeutic application in drug/gene delivery. microbubble emissions. PMI was compared with post-FUS dynamic Furthermore, in-vivo and tumor imaging performance of synthesized contrast enhanced-magnetic resonance imaging (DCE-MRI) to correl- bubbles was well explained by acoustic property measurements and ate microbubble activity with BBB opening. shell elastic and viscous parameters. Possible correlation between phys- RESULTS PMI overcomes the poor axial resolution concerns of previ- ical/acoustical properties and in-vivo/tumor imaging performance was ous passive imaging methods and provides detailed maps of micro- revealed in this study, and could be of help in future design and prac- bubble activity throughout the focal volume of the FUS transducer tical application of ultrasound contrast agents. during the BBB opening treatment. PMI is able to provide the de- tailed structure of the brain microvasculature in manner similar to power Doppler imaging, although only within the focal volume of O79 the FUS. Simplistically, PMI reveals the heterogeneous distribution of Pulse inversion based broadband subharmonic cavitation imaging microbubble activity within focal area that can be used to predict for monitoring high intensity focused ultrasound therapy where BBB opening is occurring. Indeed, post-FUS DCE-MRI images H. Zhong, X. Ma, M. Wan correlated with the distribution of microbubble activity by showing Biomedical Engineering, Xi'an Jiao Tong University, Xi'an, China contrast leakage into the brain in areas of microbubble activity. In Correspondence: H. Zhong addition to the detailed microbubble maps, PMI provides important Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O79 temporal information about the microbubble infusion kinetics and persistence within the focal volume of the FUS. OBJECTIVES This paper proposed a cavitation imaging method by CONCLUSIONS PMI can address a key technological limitation of extracting broadband subharmonic based on pulse inversion tech- poor axial image resolution with current passive cavitation imaging nique during high intensity focused ultrasound (HIFU) therapy to ob- techniques and provide a method capable of monitoring BBB open- tain monitoring images with high cavitation-to-tissue ratio (CTR), ing with FUS and microbubbles. Limitations of this new technique high cavitation detection sensitivity and high resolution. will be discussed along with its similarities and differences with exist- METHODS HIFU delivery is briefly interrupted with the HIFU com- ing methodologies. This new USgFUS modality is not only a promis- pletely off for transmission of a pair of inverted pulses (4.6 MHz) and ing technique to be used for treatment of CNS diseases, but also for acquisition of the backscattered signals from tissue samples during oncological, cardiovascular, and other medical conditions that utilize HIFU treatment. After summing the echoes of positive and negative FUS in combination with microbubbles. pulses, the subharmonic filters are designed with the center fre- quency of 2.3 MHz and the bandwidth of 20%, 60%, 80%, 100% and 140% to obtain the cavitation images. For comparison, the second O78 harmonic images were also obtained. A threshold is used to assess Long-circulating behaviour and acoustic characterization of the cavitation detection sensitivity of different cavitation images. lipid-coated microbubbles Cavitation bubble areas could be calculated by counting the number Y. Yang, H. Li, X. Guo, D. Zhang, J. Tu of pixels whose values are greater than the threshold. The larger is Nanjing University, Nanjing, Jiangsu, China the cavitation bubble area, the higher is the cavitation detection sen- Correspondence: Y. Yang sitivity. Cavitation-to-tissue ratio (CTR) could be defined as the ratio Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O78 of the intensity value averaged in the regions of interest (ROI) of Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 53 of 122 HIFU treated area to that in the ROI of HIFU untreated area. We used CONCLUSIONS These results provide more complete framework for a the normalized −6 dB width of the autocorrelation function of the large body of experiments in multiple preparations across the field of enveloped RF signal to quantify the resolution of cavitation images. US neuromodulation, lending further support to the hypothesis that RESULTS The experiments with porcine muscle demonstrated that intramembrane cavitation is responsible for ultrasonic neuromodula- the cavitation images obtained by using 80%~100% bandwidth sub- tion. They could thus pave the way towards new CNS therapeutic harmonic filters have the greatest cavitation detection sensitivity. protocols, using the only method that currently allows targeted non- The cavitation bubble areas of broadband subharmonic images invasive neuromodulation with millimeter spatial resolution essen- (1.15~3.45 MHz, including 1/2, 1/3, 1/4 subharmonic components) tially anywhere in the brain. are 1.6~2.7 times of those of narrowband subharmonic images (2.07~2.53 MHz, including 1/2 subharmonic component) with the suitable threshold. The difference of the CTR values between the O81 broadband and narrowband subharmonic images is not very large, Non-invasive high frequency transcranial focusing with a single but the CTR values of both broadband and narrowband subharmonic element transducer: experimental validation of adaptative images are much greater than those of second harmonic images. It focusing through human skulls with a 3D printed acoustic lens 2,1 2 2 2 2 was found that the resolution of broadband subharmonic images G. Maimbourg , A. Houdouin , T. Deffieux , M. Tanter , J. Aubry 1 2 was improved up to about 2 times compared with narrowband sub- Université Paris Diderot, Paris, France; Institut Langevin, ESPCI Paris, harmonic images. CNRS UMR7587, INSERM U 979, Paris, France CONCLUSIONS The proposed broadband subharmonic cavitation im- Correspondence: G. Maimbourg aging method could obtain much higher CTR value than second har- Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O81 monic imaging method. And the method has higher cavitation detection sensitivity and resolution than normal narrowband subhar- OBJECTIVES Transcranial ultrasonic brain therapy requires a wave monic imaging method. front shaping device to compensate skull- induced aberrations. Up to now, this was done with a multi-elements probe: the phase of the signal emitted by each individual transducer is adjusted in order to O80 compensate for the delay caused by the crossing of the skull. Histor- A generalized theoretical framework for understanding ultrasonic ically, a growing number of elements was used (64 elements in 2000 neuromodulation mechanisms [1], 300 in 2003 [2], 1024 in 2012 [3]) to improve the focusing. We M. Plaksin, S. Shoham, E. Kimmel propose to radically change the paradigm by achieving adaptive Faculty of Biomedical Engineering & Russell Berrie Nanotechnology transcranial focusing with a single- element covered with a 3D Institute, Technion – Israel Institute of Technology, Haifa, Israel silicone acoustic lens of variable and controlled thickness [4]. Similar Correspondence: M. Plaksin lenses have been introduced in the past to perform single or Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O80 multiple focusing patterns in homogenous propagating media [4,5] but recent 3d printing and milling capabilities make tailor-made 3D OBJECTIVES Low intensity US can noninvasively suppress or excite lenses a feasible option for transcranial adaptive focusing. The char- central nervous system (CNS) activity. Recently, we introduced the acteristic length of variation in thickness for the acoustic lens is less Neuronal Intramembrane Cavitation Excitation (NICE) framework to than 1mm. A 6cm radius lens- corrected single-element is thus explain the observed aspects of ultrasonic neuromodulation, through equivalent to an 11,000 element transducer in terms of phase shap- intramembrane space-related capacitance variations leading to cell- ing capabilities. type-selective activation effects. Here we expanded the framework to METHODS The lens is made of silicone (Elite double 8, Zhermack dissect also the impact of acoustic radiation pressure (ARP) - induced Spa, Italy). The speed of sound is c =1000m/s in silicone and silicone membrane capacitance changes on neuronal activity, as well as to c =1485m/s in water. This difference can be used to modify the water explore the effect of cell environment effective viscosity on NICE wave phase by controlling the local thickness of the lens, and thus model-related neural response. correct skull-induced aberrations [6]. The study was conducted on three METHODS We analyzed the relevant experimental literature using human skulls. The skulls were harvested and cleaned at the Saints- two sources of ARP gradients responsible for membrane dynamics: 1) Pères Anatomy Institute (Paris Descartes University) for transcranial ARP caused by ultrasonic field inhomogeneity and 2) ARP caused by ultrasound focusing studies. A 3D simulation based on computed- acoustic impedance mismatch. In addition, live brain tissue ARP- tomography of the skulls was then performed to estimate the phase gradients-subjected areal strains were evaluated in a viscoelastic shift induced by the skull at surface of the transducer (single element, brain model. In the context of the NICE model dynamics, the modi- 59 mm radius of curvature, f-number of 1, operated at 914 kHz). The fied Rayleigh–Plesset-based intramembrane cavitation biomechanics thickness of the lens was then adjusted to compensate the shifts by was calculated with cell environment viscosity higher than water vis- casting the silicone in a 3D-printed mold. The acoustic-lens-covered cosity (water viscosity was used in our previous studies), to express transducer was then immersed in water (Fig. 1). Lastly, the skull was po- the exact biological properties of cells. By coupling these biomechan- sitioned in front of the transducer with a 3D printed holder. The quality ical models to biophysical membrane models we predict dynamical of the focusing through the skull was then assessed using a 3D scan of biophysical responses of artificial bilayer membranes, and of com- the pressure field with a needle hydrophone (HNA-0400, Onda Corp., mon neocortical single cell Hodgkin-Huxley type models. Sunnyvale, CA, USA). RESULTS The augmented viscosity conditions in the NICE model lead RESULTS Figure 2 shows the acoustic pressure obtained experimentally to parabolic relationship between US frequency and threshold inten- with skull A. Axial and transverse views are presented in the absence of sities for cortical pyramidal neuron stimulation, wherein below 1 MHz skull (left), through the skull without correction (center) and with the the dependence on US frequency is negligible (with complete agree- acoustic lens (right). The lens qualitatively restores the focusing. Table 1 ment to our previous studies). The new results were found to explain reports the quantitative outcomes for the three skulls noted A, B and C. and predict the experimental results of Ye et al., Ultrasound Med Biol. The lens increased the maximum acoustic intensity by 97 ± 56 %. Com- 2016 that also captured the same behavior in their mouse study. The pared to the reference in water with no skull in place, the mean -3dB emergence of ARP gradients-induced membrane capacitance variations volume of the focus increased by 472 ± 231 % when crossing the skull, associated with membrane area changes fully explain artificial mem- whereas it only increased by 86 ± 29 % with the lens- based correction. brane experimental results and may also be responsible for neural exci- CONCLUSIONS We demonstrated that in 3D printing can be used to tation at in- vitro experimental conditions. However, these capacitance create custom-made acoustic lenses for CT- based non invasive skull- changes were found to be highly unlikely sources for neural excitation aberrations correction, making it possible to compensate for skull at in-vivo experimental conditions, when considering the areal strains aberrations without a multi- element device. Due to its simplicity, the expected to form in brain tissue during normal sonication. acoustic lens is expected to be a cheaper and less cumbersome Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 54 of 122 solution than current multi-element probes. The ultrasonic absorp- Table 1 (abstract O81). Quantitative results for assesing the efficiency tion of the lens material remains to be optimized for thermal treat- of the acoustic lens. Outcomes were obtained for three skulls named A, ments, but the setup presented here could be of interest for low B and C energy treatments such as neuromodulation or blood brain barrier opening. Acknowledgements This work was supported by the Bettencourt Schueller Foundation and the "Agence Nationale de la Recherche" under the program “Future Investments” with the reference ANR-10-EQPX-15. O82 Correlation of the lesion size in histology and mr images of the References pig brain tissue by transcranial MR-guided focused ultrasound [1] Clement G et al, A hemisphere array for non-invasive ultrasound brain 1,2 3 4 5 4 1,4 D. Paeng ,Z. Xu , J. Snell , A. H. Quigg , M. Eames , C. Jin , therapy and surgery. Phys Med Biol, 2000 [2] Pernot M et al., High power 3 3 6 4 A. C. Everstine , J. P. Sheehan , B. S. Lopes , N. Kassell transcranial beam steering for ultrasonic brain therapy. Phys Med Biol, 2003 Ocean System Engineering, Jeju National University, Jeju, Jeju, Korea; [3] Jeanmonod D et al, Transcranial magnetic resonance imaging-guided Radiation Oncology, University of Virginia, Charlottesville, Virginia, USA; focused ultrasound: noninvasive central lateral thalamotomy for chronic Neurosurgery, University of Virginia, Charlottesville, Virginia, USA; neuropathic pain. Neurosurg Focus, 2012 4 5 Focused Ultrasound Foundation, Charlottesville, Virginia, USA; Medical [4] Fjield T et al, Low-profile lenses for ultrasound surgery. Phys Med Biol, 1999 School, Virginia Common University, Richmond, Virginia, USA; [5] Melde K et al, Holograms for acoustics. Nature, 2016 Pathology, University of Virginia, Charlottesville, Virginia, USA [6] Patent: FR 1556217, July 2015 Correspondence: D. Paeng Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O82 OBJECTIVES This study is to investigate the correlation of the lesion size in histology and MR images of the pig brain tissue, which was formed by a transcranial magnetic resonance-guided focused ultra- sound (tMRgFUS) system and observed in 3 days after sonication. The lesion was generated by relatively low temperature between 46 ~ 52°C for an appropriate time to reach a target thermal dose in CEM (cumulated equivalent minutes). The target thermal dose was below 200 CEM which was obtained by variation of pulse duration of the tMRgFUS system for constant target temperature through a closed-loop system based on MR thermometry. METHODS A tMRgFUS system (ExAblate 4000 Neuro 650 kHz system, InSightec, Israel) was used for this pig experiment. An MRI system Fig. 1 (abstract O81). (A) Experimental setup in the water tank (Discovery MR75-3.0T, GE Medical systems) was used for thermom- during the 3D-scan of the pressure field by the needle hydrophone. etry and pre- and post-imaging. A closed-loop control system was The skull is supported on the internally-designed holder by two implemented on a personal computer to control pulse duration of elastic bands. (B) an axial cut of the setup. The acoustic lens covers the tMRgFUS system at a certain acoustic power in order to maintain the surface of the transducer a target temperature based on the MR thermometry. Temperature distribution and accumulated thermal dose in the target brain tissue was calculated every 3.7 seconds. Sonication was stopped when a prescribed thermal dose was delivered to the targeted tissue. The proportionate and integral coefficients of the PI controller were found to be 5 and 1.5, respectively, from several phantom experi- ments and first acute pig experiment while the acoustic power was varied up to a few hundred watts. Twelve chronic pig experiments with craniectomy were conducted, and post MR images within 1 hour and at 3 days were taken after sonication. Brain tissue was har- vested in 3 days before euthanasia for histology. Four lesions were generated on both sides of the thalamus of each pig. Temperature in the pig brain tissue was estimated by rectal temperature for the MR thermometry baseline. This study was approved by the University of Virginia Institutional Animal Care and Use Committee (ACUC). RESULTS Among 12 chronic pigs, one pig had a problem with intro- duction of air bubbles during surgery procedure and another one had an ACUC issue, so that these data could not be used. Three other pigs had a lower rectal temperature below 32°C so that the thermal dose computation was not reliable, leading to exclusion of Fig. 2 (abstract O81). Acoustic pressure acquired by a needle the data analysis. No obvious lesions were observed in MR images hydrophone. From left to right: the reference focusing (no skull and taken in an hour of sonication for thermal dose less than 200 CEM. no lens), the focusing through a human skull without any correction Figure 1 shows the lesion diameters measured in MR T2-weighted and the focusing through a human skull with the custom-made axial images and histology in 3 days of sonication as a function of acoustic lens. The black crosses depict the position of the reference thermal dose in CEM. There are lesions in all MR images and hist- focusing ology over 77 CEM except 101 CEM where no lesion is shown in Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 55 of 122 histology while MR shows a lesion. The lesion diameters from MR im- the DMUA behind the skull sample (Fig. 1). In other experiments, skull ages and histology are mostly 3 ± 1mm for thermal dose over 77 samples were embedded in a tissue- mimicking phantom and posi- CEM. There are 6 lesions in MR T2-weighted axial image but no le- tioned at a distance of approximately 32-mm from the apex of the sion was found in histology, whose thermal doses were 23, 30, 46, DMUA (corresponding to the skull position during in vivo experiments.) 53, 69, 101 CEM. Below 75 CEM, no lesion was found except one in A thermocouple was inserted so that its junction was closest to the geo- 18 CEM in histology. One lesion on 18 CEM is shown in both MR metric focus. Two modes of imaging were used: 1) Synthetic-aperture image and histology, and the rectal temperature was low to 33.3°C. imaging, which provided larger field of view to guide the placement of Lesion diameter observed in histology (y) is highly correlated with the tFUS beam, and 2) Single-transmit focus (STF) imaging, which one in MR T2-weighted axial imaging (x), and their function and cor- allowed for the characterization of the tFUS beam interaction with the relation coefficient are y=0.90x and r =0.66, respectively. Even skull. SA images were used to place target point(s) and critical point(s) though the histology is sliced in coronal plane, the histology lesion used by the refocusing algorithm. Raw echo data received by each array diameter is highly correlated with MR axial diameter. element were used to form data matrices corresponding to the target CONCLUSIONS The lesions were formed in the pig brain tissue in 3 and critical point(s). These matrices were used to solve an optimal re- days of sonication by tMRgFUS for thermal dose over 77 CEM, except focusing problem based on the concept of propagation operators from one in histology for thermal dose of 101 CEM. The diameter of the le- the array elements to the target point(s). The refocusing algorithm ran sions was measured to be mostly 3 ± 1mm in both MR T2-weighted automatically once the user specified the target and critical point. It was axial images and histology, which is close to focal size. The diameter implemented on a software- defined ultrasound (SDUS) architecture that in histology is 10 % smaller than the one in MR T2-weighted axial im- allowed the real-time computation of the refocused excitation vector ages with correlation coefficient of 0.66. There are some differences and immediate download to the driver within milliseconds. To demon- in lesions between histology and MR images, and further systematic strate the improvement in focusing gain, we have analyzed the change researches are required for the reasons. in echogencity from the target when insonified using geometric and re- focused STF imaging beams. In addition, the temperature rise due to therapeutic tFUS beams with geometric focusing and optimal refocusing were directly measured. The heating rates were computed by taking the time derivative of the measured temperature profiles. RESULTS Figure 1 also shows STF images generated from uniform elem- ent excitation (geometric focusing) and refocused excitation vector. The refocused STF image exhibits increased echogenicity from the target (thermocouple) and reduced echogenicity from the skull, where a critical point was located near the medial suture line. This is more clearly shown by the line graphs in Figs. 2 and 3. To demonstrate the increased thera- peutic gain due to refocusing, the geometric focusing beam was used in therapeutic mode to produce heating at the thermocouple for 10 second duration followed by a refocused beam. Figure 4 shows the heating rates produced by the geometric focus and the refocused beam as estimated from the thermocouple measurements. It is quite clear that the refocus- ing gain was increased approximately by a factor of 3. Specifically, the heating rate due to the refocused beam was 3.636 C/sec while that of the geometrically focused beam was 1.179 C/sec. This result was typical, but variation in the refocusing gain were observed depending on which part of the skull was traversed by the tFUS beam. CONCLUSIONS The results shown above demonstrate the feasibility of Fig. 1 (abstract O82). Comparison of the diameter of MR using STF imaging data for characterizing the quality of the tFUS beams T2-weighted axial image with the one of histology as a function noninvasively. More importantly, they demonstrate the feasibility of re- of thermal dose in CEM focusing the beam to provide significant improvement in focusing gain as evidenced by the 3-fold increase in heating rate shown. The result shown here was typical of numerous experiments where the tFUS O83 beam traversed different regions within the skull with different distor- Real-time image-based transcranial refocusing of dual-mode ultrasound arrays: ex vivo results tion. In every case, however, a significant improvement in focusing gain was achieved. The results shows were obtained using a 1D array and H. Aldiabat, P. D. O'Brien, D. Liu, E. S. Ebbini the refocusing was achieved only in the axial lateral directions. A 2D Electrical Engineering, University of Minnesota, Minneapolis, Minnesota, array would have produced even higher refocusing gain by taking ad- USA Correspondence: H. Aldiabat vantage of the elevation direction. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O83 OBJECTIVES Transcranial focused ultrasound (tFUS) is gaining wider ac- ceptance in a range of therapeutic applications for the treatment of brain disorders. A major challenge towards widespread use of tFUS- based therapies stems from the complexity of the skull that could result in severe loss of focusing gain. Using extensive transskull hydrophone scans in rodent and human skulls, we have documented a range of tFUS beam distortions from mild shifting to complete splitting. These distortions could compromise the specificity of targeting brain circuitry and/or cause collateral damage at unintended locations. In this paper, we present first ex vivo results demonstrating the feasibility of image- based refocusing to regain the focusing gain. METHODS A 3.5-MHz dual-mode ultrasound array (DMUA) prototype (64 elements, concave with 40-mm radius of curvature) was used. The DMUA was used to image and generate therapeutic focused ultrasound beams through rodent skull samples ex vivo. In some experiments, the Fig. 1 (abstract O83). See text for description target was the nose of a cone positioned near the geometric focus of Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 56 of 122 correction) acquired using the array elements as receivers. Following aberration correction, hydrophone measurements of the focal pres- sure amplitudes and focal beam profiles were acquired, as were mea- surements of the focal pressure amplitudes as a function of the electronic steering distance through the skull. Lesioning experiments were carried out in red blood cell tissue phantoms to compare how the different aberration correction modalities affected lesioning. Fi- nally, volumetric treatments of large clots through the skull were conducted and treatment efficacy for each aberration correction case was evaluated as a function of clot volumes liquefied. RESULTS Aberration correction performed based on hydrophone Fig. 2 (abstract O83). See text for description measurements and backscatter methods increased the peak negative pressure through the skull at the focus by 90% and 60%, respect- ively, compared to the no aberration correction case. The electronic steering radius at which focal pressures above the nucleation thresh- old could be reached improved by more than 100% with aberration correction, increasing from 7 mm without aberration correction to over 15 mm with aberration correction. The cavitation clouds and le- sions generated in both aberration correction cases were equal in size and up to 30% larger than the no aberration correction case using the same input power to the ultrasound array due to increased focal pressure amplitudes. The increased pressure amplitudes and steering ranges in the aberration correction cases resulted in signifi- cantly improved transcranial clot liquefaction speeds compared to the no aberration correction case. CONCLUSIONS This study indicates that all measures of histotripsy applied transcranially benefit from aberration correction. While hydrophone aberration correction was seen to provide the greatest increases in focal pressure, steering range, and liquefaction rates, ab- erration correction applied based on backscatter methods was seen to perform comparably to the hydrophone case and provide signifi- cant improvements over the no aberration correction case. These re- sults indicate that backscatter aberration correction can offer a non- invasive method of improving the efficacy of transcranial histotripsy comparable to hydrophone based correction methods. Fig. 3 (abstract O83). See text for description O85 O84 accumulated thermal dose in guiding essential tremor treatment: A comparative study of transcranial histotripsy applied with and repeated sonications with peak temperatures below 55°C without aberration correction 1 1,2 3 3 1,2 Y. Huang , R. M. Jones , N. Lipsman , M. L. Schwartz , K. Hynynen J. R. Sukovich, T. Gerhardson, T. Hall, J. Macoskey, C. A. Cain, Z. Xu 1 2 Sunnybrook Research Institute, Toronto, Ontario, Canada; Department Biomedical Engineering, University of Michigan, Ann Arbor, Michigan, of Medical Biophysics, University of Toronto, Toronto, Ontario, Canada; USA Division of Neurosurgery, Sunnybrook Health Sciences Centre, Toronto, Correspondence: J. R. Sukovich Ontario, Canada Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O84 Correspondence: J. R. Sukovich Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O85 OBJECTIVES Histotripsy is an ultrasound therapy that generates cavi- tation bubble clouds to fractionate soft tissue using short duration, OBJECTIVES Magnetic resonance guided focused ultrasound has re- high-amplitude ultrasound pulses. Previous studies have shown that cently been approved by the FDA for the treatment of essential histotripsy is capable of producing lesions through the skullcap with tremor. During these treatments, peak temperatures between 55 and or without using aberration correction. Transcranial ultrasound ther- 60°C are generally desired to produce reliable lesions at the target in mal therapy cannot treat within 2 cm of the skullcap and cannot the thalamus. However, in some patients a peak temperature of 55°C treat large volumes in the brain due to skull heating, which makes it cannot not be reached, either because of their skull properties or unsuitable for brain tumor treatments. In comparison, histotripsy can due to a low tolerance to the pain associated with substantial skull be used to treat targets near the skullcap as well as large volume tar- heating. In these patients, repeated sonications with peak tempera- gets transcranially with minimal skull heating by using low duty cycle tures from 50 to 54°C were typically achievable. The accumulated ef- pulses (<0.1%). This study presents comparative analyses of the pres- fects of these sonications may also result in a sufficient thermal dose sure fields, cavitation clouds and lesions, and treatment rates of his- to produce a lesion. Therefore, establishing a proper threshold for totripsy applied transcranially with and without aberration correction. the accumulated thermal dose (ATD) is important for guiding these METHODS Histotripsy pulses were delivered through ex-vivo human repeated sonications with lower peak temperatures. In this study, skullcaps mounted centrally within a 500 kHz, 256-element histo- clinical treatments at our institution with maximum peak tempera- tripsy transducer with transmit-receive capable elements. Aberration tures below 55°C were retrospectively reviewed. ATD maps were cal- correction was applied in two ways, 1) based on hydrophone mea- culated offline with corrections for chemical-shift artifacts and were surements and 2) based on histotripsy pulse-backscatter measure- correlated to the lesion sizes observed at follow-up imaging to find ments (from bubbles generated transcranially without aberration Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 57 of 122 the best estimate for the ATD threshold in terms of cumulative artery occlusion (dMCAO), and used both behavior (neurological se- equivalent minutes at 43°C (CEM ). verity score) and histological measures were used to assess the neu- METHODS From Jul 2015 to Jan 2017, 28 patients with medication- roprotection effects of TUS on ischemic stroke rats. In the second refractory essential tremor have been treated with a clinical focused study, we applied TUS stimulation as a preconditioning to rats before ultrasound system (ExAblate Neuro, 650 kHz central frequency, ischemia induced by photothrombosis, and both cerebral blood flow InSightec, Tirat Carmel, Israel) and a 3 T MR scanner (MR750, GE and histological measures was used to assess the neuroprotection ef- Healthcare, Milwaukee, WI, USA). Among these, 5 patients were fects of TUS preconditioning on ischemic stroke injury. treated with a maximum peak temperature below 55°C (53 or 54°C). RESULTS In the first study, the rats received TUS after dMCAO For the ATD calculation over multiple sonications, chemical-shift arti- showed significantly lower NSS (n=10, 5.5±2.5) than the Control facts in MR thermometry were corrected retrospectively in Matlab if group (n=10, 10.5±1.4) (p<0.01). In addition, the stroke rats received the misalignment between successive sonications exceeded 1 mm TUS had significantly reduced cortical infarct volume than the control by manually shifting the centre of the heating area along the group (13.78% ± 8.18%, n=16, versus 43.39%±2.33%, n=16, p<0.01). frequency-encoding direction. The ATD was then integrated in the What’s more, Immunohistochemical staining indicated reduction of axial plane using the standard thermal dose model. The spatial di- neutrophils in the affected area, and laser speckle imaging showed mensions of the ATD at 17, 40 and 240 CEM were measured and significant increase of a cerebral blood flow after TUS, which consist- correlated to the lesion sizes measured in T1- (3D FSPGR, TR 8.3 ms, ently supported the neuroprotection of pTUS in ischemic brain injury. TE 3.3 ms, slice thickness 1.2 mm) and T2- (FRFSE, TR 5200 ms, TE In the second study, the ischemic stroke rats received TUS precondi- 100 ms, slice thickness 3 mm) weighted images acquired on the first tioning had smaller ischemic areas during stroke induction, and 24 day follow-up. Logarithmic regression was applied on the correlation and 48 hours after the stroke, and smaller infarct volume (1.770 ± coefficients to find the best estimates for the thermal dose 0.169%) than the controls (3.215 ± 0.401%) (p<0.01). Moreover, the thresholds. stroke rats with TUS preconditioning experienced lower volume of RESULTS Thermal lesions were successfully produced in all 5 pa- brain edema than the control group. tients. The lesion size in the axial plane (AP and LR) as measured on CONCLUSIONS In the first study, both behavior and histological re- T2w images (5.0±1.9 mm) was 16% larger than that found on the sults suggested that TUS on ischemic core immediately after ische- corresponding T1w scans (4.3±2.0mm), the latter of which primarily mic stroke could be neuroprotective. In the second study, the results revealed vascular damage. As a result, logarithmic regression showed demonstrated that TUS preconditioning before photothrombosis that the best correlation of the ATD to the lesion size on T2w im- could provide neuroprotection by increasing the brain’s tolerance to aging was close to 100 CEM (linear regression slope=0.98, R = subsequently induced focal ischemic injury. 0.62), whereas the best correlation of the ATD to the lesion size on T1w imaging was close to 200 CEM (linear regression slope=0.94, R = 0.84). 17 and 40 CEM significantly overestimated the lesion size Posters in T1w images by 50% and 30%, respectively. These results are in good agreement with our previous study, which included 36 P1 treatments with peak temperatures mostly above 55°C. Therapeutic effects of chemo-agent delivery across the CONCLUSIONS An ATD of 100 CEM showed a good correlation to blood-brain barrier using mrgfus: in vivo study using the lesion size measured on T2w images acquired one day post treat- drug-resistant breast cancer brain metastasis model ment. This threshold appears to be valid both for lesions produced Eun-Joo Park, Yuri Cheon, Yun Deok Ahn, Jae Young Lee by sonications above 55°C, and by repeated sonications below 55°C. Radiology, Seoul National University Hospital, Seoul, Korea These results provide important guidance for predicting the lesion Correspondence: Eun-Joo Park size induced during focused ultrasound-based treatment of essential Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P1 tremor. (No Image Selected). OBJECTIVES Several studies have shown the promising results of drug delivery across the BBB using MRgFUS in combination with O86 microbubbles. However, more studies are required to evaluate this Neuroprotection of low-intensity transcranial ultrasound treatment technique prior to apply in clinic. This study was designed stimulation in ischemic stroke rats to evaluate the therapeutic effects of chemo-agent for brain J. Sun, H. Li, S. Tong, metastasis cancer by delivering drug across the BBB using MRgFUS School of Biomedical Engineering, Shanghai Jiao Tong University, andmicrobubbles. Shanghai, China METHODS In vivo studies were performed in two steps. For the first Correspondence: J. Sun step, multiple BBB-openings were performed at four different FUS Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O86 condition along with microbubbles (SonoVue) to find safe and effect- ive treatment condition. The treatment was performed once a week OBJECTIVES Transcranial Ultrasound Stimulation (TUS) is a non- for six weeks. In the second step, breast cancer cells (BT-474) were invasive neuromodulation technique which modulates the neural ac- pre-treated with chemo-agent prior to the inoculation in the rat brain tivity by delivering specific pulsed ultrasonic waves through intact for the brain metastasis model. Animals were treated in three groups: scalp and skull to the targeted brain regions. In recent years, studies control, chemo-agent treatment only, and chemo-agent treatment have demonstrated that low-intensity TUS is a safe neuromodulation with BBB-opening. FUS condition and injection volume of microbub- technique with high focality and deep penetration depth, and has bles for BBB-opening were obtained from the first step experiment. therapeutic effects for brain diseases like epilepsy and stroke in ani- Animals were treated on a weekly basis for six weeks and post-treat- mal models. With these advantages, we have applied TUS in precon- ment tumor growth monitoring was followed for 12 weeks. ditioning and neuroprotection of ischemic rats, so as to verify the RESULTS As restuls of this study, histological evaluation of each BBB- neuroprotection effects of TUS in stroke rats. opening FUS condition in combination with SonoVue, therapeutic ef- METHODS In the first study, we applied a low-intensity TUS to the is- fects of chemo-agent treatment with BBB-opening, and the survival chemic cortex of SD rats immediately after a distal middle cerebral rate of each treatment will be presented. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 58 of 122 CONCLUSIONS This study evaluated the FUS condition for multiple P3 BBB-opening in combination with microbubbles, SonoVue. With the Focused ultrasound-enabled delivery of radiolabeled nanoclusters treatment conditions, therapeutic effects of chemo-agent delivery to the pons with concurrent pet imaging 1 4 4 4 across the BBB for brain metastasis cancer model by using breast Dezhuang Ye , Sultan Deborah , Hannah Luehmann , Yuanchun Tai , 2 4 3, 5 cancer cells pre-treated chemo-agent. Josh Rubin , Yongjian Liu , Hong Chen Mechanical Engineering and Material Science, Washington University in St. Louis, Saint Louis, Missouri, USA; Pediatrics Hematology/Oncology, P2 WashingtonUniversity in St. Louis, Saint Louis, Missouri, USA; Biomedical Optimizing passive cavitation mapping by refined minimum Engineering, Washington University in St. Louis, Saint Louis, Missouri, variance-based beamforming method: performance evaluationsin USA; Radiology - Rad Sciences, Washington University in St. Louis, Saint macaque models for blood-brain barrier disruption Louis, Missouri, USA; Radiation Oncology, Washington University in St. 1,2 1 2 1 2 Tao Sun , Calum Crake , Brian Tracey , Costas Arvanitis , Eric Miller , Louis, SaintLouis, Missouri, USA Nathan McDannold Correspondence: Dezhuang Ye Radiology, Brigham and Women's Hospital; Harvard Medical School, Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P3 Boston, Massachusetts, USA; Electrical and Computer Engineering, Tufts University, Medford, Massachusetts, USA OBJECTIVES Pontine glioma is the single greatest cause of brain Correspondence: Tao Sun tumor-related death in children. Pons is a major structure in the Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P2 upper part of the brainstem, which is involved in the control of func- tions such as breathing, hearing, taste, balance, and communication OBJECTIVES Microbubble-mediated focused ultrasound (FUS) therapies between different parts of the brain. The unique location of the pons harness mechanical and/or thermal effects to deliver drugs or ablate precludes surgical interventions while conventional chemotherapy tissues. Passive acoustic mapping (PAM) enables the spatio-temporal shows little improvement in survival because the blood-brain barrier monitoring of cavitation activity, which is critical for the clinical transla- (BBB) remains intact in pontine glioma. Recent efforts have focused tion of this technique. Traditional PAM is based ondelay-and-sum (DAS) on improvements in chemotherapeutic agents employed and in beamforming, a method whose quality tends to deteriorate due to is- methods of localized and targeted drug delivery. Several drug deliv- sues including multi-bubble interference, distortion in the wavefront ery strategies, such as convection-enhanced delivery and intranasal caused bythe presence of the skull, unmodeled variability of array ele- delivery, have been proposed to bypass the BBB for the treatment of ments, etc. To provide for robustness, here we consider the use of mini- pontine glioma but have met withminimal success. Focused ultra- mum variance adaptive beamforming toPAM and demonstrate sound (FUS) combined with microbubbles has been successfully used significant improvement in image quality compared to DAS. in the noninvasive and localized trans-BBB delivery of various drugs METHODS Experiments were performed in phantom with macaque and several nanoparticles for the treatment of brain cancer. 64Cu- skull and in vivo for monitoring FUS-induced blood-brain barrier disrup- alloyed gold nanoclusters (64CuAuNCs) composed of a few gold tion in a clinical MRIguided FUS system (ExAblate 4000, InSightec, Haifa, atoms and radiolabeled copper atoms are emerging theranostic Israel), which was integrated with a clinical 3T MRI unit (GE Healthcare). agents for cancer treatment. Their small sizes with PET imaging A 1024-element phased array was driven to transmit 10-ms pulsed FUS capability present unique advantages to be integrated with the FUS- at 220 kHz. RF data for PAM were acquired using a research ultrasound technique. The goal of this study was to demonstrate that FUS can imaging system (Verasonics, Redmond, WA) passively. The imaging enable the delivery of 64CuAuNCs noninvasively and locally to the probe (L382, Acuson, WA) was a 128-element linear array (fc = 3.21 pons with concurrent PET imaging capability for quantitative evalu- MHz; bandwidth: ~ 75%). Transition and receiving systems were syn- ation of the delivery outcome. The long-term perspective is to apply chronized and the first180 μs of RF-data were recorded for each burst. this novel image-guided drug delivery platform for the treatment of RESULTS Minimum variance distortionless response (MVDR) method pontine gliomas. was evaluated and further improved by adding diagonal loading and METHODS First, six wild-type mice were treated by FUS for evaluat- using subarray for covariance estimate. Results demonstrate the reso- ing the feasibility of safe and localized FUS drug delivery to the pons lution improvement and contrast enhancement using both trad- using a model drug: 40 kDafluorescently-labeled dextran with a itional and refined MVDR Beamformers compared toDAS. Axial full hydrodynamic diameter (~2.3 nm) that was close to that of the width at half maximum was tightened down to 79.5% and 38.5% of 64CuAuNCs (3-5 nm). Dextran delivery outcomes were evaluated that in DAS image for traditional and refined MVDR Beamformers, based on fluorescence images of ex vivo brain slices and safety was respectively. Moreover, refined MVDR method greatly enhances the assessed using histological staining. Second, another group of six robustness when traditional MVDR Beamforming induces more mice was used for demonstrating the feasibility of FUS-enabled artifacts due to the self-nulling effects (shown in Fig. 1). 64CuAuNCs delivery to the pons. FUS sonicated the left pons in the CONCLUSIONS It was demonstrated that the refined MVDR method presence of systemically administrated microbubbles. After FUS son- improved the image quality significantly compared to traditional DAS ication, mice were transferred to microPET/CT facility. 64CuAuNCs in method. We anticipate that the proposed methods will improve our 100 μL saline was injected into the mice via the tail vein, followed by ability to monitor and control FUS-induced cavitation-based therapies. PET scanning of themice at 1 hr, 4 hr, and 24 hr. To further validate the delivery of 64CuAuNCs, auto-radiography of ex vivo brain slices was performed, followed by inductively coupled plasma mass spec- trometry (ICP-MS) quantification of the gold concentration in the brain. At last, the biodistribution of 64CuAuNCs was evaluated using gamma counting and ICP-MS. RESULTS Fluorescence images of mouse brain slices showed localized delivery of the dextran at the FUS targeted left side of the pons. Hematoxylin and eosin stainingof the whole brain confirmed that the treatment did not cause histological level tissue damage. PET images showed targeted delivery of 64CuAuNCs to the pons and quantitative analysis of the PET images found the delivery efficiency was1.44%ID/g. Auto-radiography further validated the successeful delivery of 64CuAuNCs to the pons. ICP-MS quantification found 3-fold increase in gold concentration at the FUS-treated left side of the pons compared Fig. 1 (abstract P2). An example to show the -3dB profile. The with the contralateral nontreated right side ofthe pons. Biodistribution presented method can enhance the robustness of traditional MVDR study showed the 64CuAuNCs was cleared by liver and kidney, demon- when the main scatters were self-nulled strating the reduced systematic toxicity of this nanoparticle. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 59 of 122 CONCLUSIONS The unique location of the pons and the intact BBB in pontine glioma present a critical need for noninvasive and localized tech- niques to overcome theBBB. Although FUS technique has been under de- velopment for brain cancer drug delivery for more than one decade, this is the first study that demonstrated the successful drug delivery to the pons. Meanwhile, nanomedicine is the next-generation platform technol- ogy for cancer therapy, and FUS has been shown promising in the deliv- ery of several nanoparticles. This study demonstrated the unique integration of the FUS technique with nanoclusters for brain drug deliv- ery. The small sizes of nanoclusters presented unique advantages in trans-BBB delivery and brain parenchyma penetration. Last but not least, contrast-enhanced MRI is currently the standard technique for FUS-in- Fig. 1 (abstract P4). Typical ESD responses and the animal brain duced BBB opening quantification. Contrast-enhanced MRI monitors con- sections of a two animals with (a) or without (b) successfully induce trast agent leakage and assumes an indirect correlation between the BBB opening. The Red circles in (a) shows the EB stained regions delivery of the contrast agent and therapeutic agents. PET imaging of which represent the BBBopened region. (d) Comparison between radiolabeled nanoparticles provides a noninvasive, highly sensitive, and the peak ESD magnitudes with the BBB status (intact or opened). quantitative method for directly evaluating the trans-BBB delivery of n = 8 for eachgroup nanoparticles. P5 P4 Microbubble-facilitated focused ultrasound enhances the delivery An acoustic emission-feedback planar ultrasound system for of virus-like particles into the brain localized blood-brain barrier opening and monitoring Chia-Jun Lin, Hong-Wei Yang, Hao-Li Liu 1 1 1 2 Yu-Xian Lin , Yu-Chien Lin , Chih-Hung Tsai , Wen-Shiang Chen , Claude Department of Electrical Engineering, Chang Gung University, Taoyuan 3 1,4 Inserra , Hao-Li Liu City, Taiwan Electrical Engineering, Chang Gung University, Taoyuan City, Taiwan; Correspondence: Chia-Jung Lin Physical Medicine & Rehabilitation, National Taiwan University Hospital, Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P5 3 4 Taipei, Taiwan; Université de Lyon, Lyon, France; Neurosurgery, Chang Gung Memorial, Taoyuan City, Taiwan OBJECTIVES Microbubble-facilitated focused ultrasound (FUS) has been Correspondence: Yu-Xian Lin applied to transiently induce blood-brain barrier (BBB) opening nonin- Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P4 vasively and locally. Although substance with moderate size has been shown its feasibility to be delivered into the brain via this technique, OBJECTIVES In this study, we proposed a novel planar ultrasound ap- large molecular delivery (MW > 500 kDa) is still challenging and the paratus design that can provide PCD and real-time analysis, with the delivered effeminacy is still unknown. In this study, we aim to test the intention to perform real-time planar ultrasound BBB opening moni- CNS delivered efficacy of a novel large molecule, virus-like particles toring and control. (VLPs;MW = 2000 kDa), into the brain via microbubble-facilitated FUS METHODS A planar ultrasound probe integrating with a lateral mode BBB opening, and investigate the penetration and distribution. transducer ring was coaxially arranged, which the former one is to METHODS A total of 12 ICR mice were employed in this study. A 400 perform energy exposure and the later one is to passively receive kHz spherically focused transducer was used to deliver ultrasound the backscattered emission and characterize the subharmonic/ultra- energy (0.155 MPa; burstlength = 10 ms, PRF = 1 Hz, duration = 60 harmonic emission spectrum density (ESD) during treatment. Invivo s). The focal zone was located at left cerebral hemisphere with 2-3 tube phantom experiments were conducted to characterize the de- mm depth. Brains were harvested at 4 hours post FUS treatment, pendence of ESD change with microbubble infusion. In-vitro experi- and the frozen samples were sectioned and observed with a fluores- ments were employed to characterize the dependence of ESD cent microscope (TissueFAX Plus, Austria). Tissue slices were then change, and in-vivo experiments were employed to characterize the stained byimmunofluorescence and observed to investigate the dis- effect to induce BBB opening. Evans blue extravasations and HE tribution of VLPs in brain cells. GFP-bounded dextrans with the size staining was conducted to provide histological confirmation. ranging from 70 – 2000 kDa were employed as another molecular RESULTS This study demonstrates the capability in using planar ultra- surrogate and compare with the VLP delivery. Evans blue (EB) was sound system to open the BBB. The ESD response well corresponds used as the indicator of BBB opening. to the animal brain section, where the EB well stained on the ultra- RESULTS FUS exposure with the presence of microbubbles were shown sound exposure position to induce successful BBB opening. For ani- to be able to locally open the BBB at the target site. VLP has shown evi- mal group with BBB still intact after exposure, the peak ESD was dence to be delivered into the brain (Fig. 1). When correlated with the observed to be 0± 3 dB. On the other hand, in animal group with observation of GFP-bounded dextrans, it was observed that 70 kDa dex- successful BBB-opening, the peak ESD was observed to be signifi- trans spread in the left hemisphere, while 2000 kDadextrans aggregated cantly higher (17± 12 dB) than the BBB-intact groups (Fig. 1). The 4- at some spots. More VLPs are distributed near the hemorrhage sites in dB ESD level was found to be a valid threshold level to well discrim- the left hemisphere after treated with microbubble-facilitated FUS, com- inate the BBB-intact and BBB-opened group. With the integration of paredto the non-hemorrhage sites. VLPs are co-localized with neuron nu- the lateral-mode ring transducer, the proposed system also demon- clei (NeuN), while not obviously with gilal fibrillary acidic protein (GFAP). strates the feasibility to monitor the BBB-opened status, making the 2000 kDa dextranaggregated between the brain cells but showed no evi- real-time control of the process become feasible. dence of the entry in brain cells. CONCLUSIONS This study may provide valuable information toward CONCLUSIONS In this study, we have demonstrated that large mo- designing a planar ultrasound treatment apparatus for the purpose lecular delivery up to 2000 kDa is feasible when combining with of BBB opening and brain drug delivery. microbubble-facilitated FUS BBBopening. We observed that the VLPs Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 60 of 122 when entering into brain is able to penetrate into neuron cells, sup- the results of western blot were calculated on amounts of P-gp com- port the VLP-cell endocytosis is occurred. VLP preserve superiority in pared with β-actin. In addition, the change of P-gp in sonicated area its high endocytosis feature, making the VLP has the potential to shows percent assuming that P-gp of non-sonicated area is100%. At serve as a gene carrier to perform CNS gene delivery. 1 hour, P-gp of sonicated region was reduced by 15% and showed the lowest reduction on 1 day (76%). From 2 to 5 day, P-gp seemed to be recover gradually. It was reduced by 61% on 2 day and 43% on 3 day. On 5 days, it was completely recovered and showed no de- crease (Fig. 1B). CONCLUSIONS The results of this study provide the information needed to take into account the dynamics P-gp change over time after FUS-induced BBB disruption. Therefore, these results could sug- gest more detailed treatment protocols when FUS-induced BBB dis- ruption treatment along with P-gp substrate drug. Fig. 1 (abstract P6). MR image (A), Pglycoprotein quantification graph (B) Fig. 1 (abstract P5). (A) After microbubblefacilitated FUS treatment P7 at left hemisphere, EB entered into the brain, indicating the BBB is Focused ultrasound-induced blood-brain barrier opening enhances opened. Some hemorrhage was observed. One unit of scale is 1 GSK-3 inhibitor delivery for amyloid-beta plaque reduction mm. (B) VLP fused with mcherry fluorescent protein entered into the 1,2 2 3 4,5 PoHung Hsu , Yi-Hsiu Chung , KunJu Lin , LiangYo Yang , left hemisphere after BBB opening induced by microbubblefacilitated 2 1,6 Tzu-Chen Yen , Hao-Li Liu FUS.VLP was shown as red. Nuclei were stained by DAPI. (C) Compared 1 2 Electrical Engineering, Chang Gung University, Taoyuan City, Taiwan; Center with saline injection or VLP injection without FUS, more VLPs were forAdvancedMolecular ImagingandTranslation, ChangGungMemorial colocalized with NeuN after BBB opening, indicated its entry into the Hospital, Taoyuan City, Taiwan; Department of Nuclear Medicine, Chang neuron cells. VLP was presented as green, NeuN was as red and nuclei Gung Memorial Hospital, Taoyuan City, Taiwan; Department of Physiology, was as blue School of Medicine, College of Medicine, China Medical University, Taichung, Taiwan; Department of Biotechnology, Asia University, Taichung, Taiwan; Medical Imaging Research Center, Institute forRadiological Research, Chang P6 Gung University and Chang Gung Memorial Hospital, Taoyuan City, Taiwan The kinetics of P-glycoprotein after blood-brain barrier Correspondence: PoHung Hsu disruption in rat brain by MRI-guided focused Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P7 ultrasound Mun Han, Danbi Kim, Sanghyun Ahn, Juyoung Park OBJECTIVES Alzheimer's disease (AD) is a chronic neurodegenerative Daegu-Gyeongbuk Medical Innovation Foundation, Daegu, Korea disease that is the leading cause of age-related dementia. Currently, Correspondence: Mun Han therapeutic agent delivery tothe CNS is a valued approach for AD Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P6 therapy. Unfortunately, CNS penetration of therapeutic agents is greatly hampered by the blood-brain barrier (BBB). Focused ultra- OBJECTIVES Multi-drug resistant proteins in brain endothelial cells sound (FUS) exposure has been demonstrated to temporally open pump drugs in brain tissues out into the blood and could diminish the BBB, thus promoting therapeutic agent delivery to the CNS for drug retention and drug efficacy. In previous our study, a representa- AD therapy. In this study, we aimed to evaluate whether the use of tive multi-drug resistant protein, P-glycoprotein (P-gp), was signifi- FUS-induced BBB opening to enhance GSK-3 inhibitor delivery can cantly reduced at 24 hrs after blood brain barrier (BBB) disruption by promote Aβ plaque clearance and synthetic regulation in a trans- focused ultrasound (FUS). In this study, we investigated the kinetics genic small animal model (Fig. 1a). of the P-gp after FUS-induced BBB disruption. METHODS FUS-induced BBB opening on APP/PS1 transgenic mice METHODS The rat brain in thalamus was sonicated (0.5~0.55 MPa) was performed unilaterally, with the contralateral hemisphere serving transracially using a 1 MHz FUS transducer with 10 ms bursts of ultra- as a reference. AV-45PET/CT imaging was attempted to detect plaque sound wave at 1Hz pulse repetition frequency for 120s, combined distribution and concentration in vivo, and autoradiography (ARG) with intravenous injection of a microbubble ultrasound contrast was conducted ex vivo to quantitatively confirm theAβ plaque reduc- agent (Definity; 400 μl/kg). T1 and T2 weighted MRI scan were used tion. Immunohistochemistry staining was also performed to confirm to guide FUS sonication into the targeted brain and confirm the BBB the GSK-3 expression level. disruption. Then, the sonicated rat brains were extracted at different RESULTS Results from AV-45 PET/CT imaging showed positive Aβ time points (1 hr, 1, 2, 3, 5day) after BBB disruption. In order to meas- plaque regulation in vivo,and ex vivo autoradiography (ARG) further ure the P-gp expression, choroid plexuses were removed from all showed significant radiolabelled tracer detectability (up to 26.72% Aβ ventricles and then western blot analysis was performed using plaque reduction). Immunohistochemistry revealed that the GSK-3 in- antibody against P-gp (1:100) and β-actin (1:2000). P-gp expression hibitor effectively blocked plaque synthesis up to60.6% in the GSK-3 sonicated area and non-sonicated area were compared. immunoreactive area, confirming the proposed therapeutic pathway. RESULTS The BBB disruption was confirmed in the thalamus region CONCLUSIONS We demonstrated that FUS-induced BBB opening to through T1 weighted images with MR contrast agent (Fig. 1A). All enhance GSK-3 inhibitor delivery can efficiently reduce amyloid-beta Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 61 of 122 plaques in transgenicmouse brains. The results showed that the use Bracco, 500 ul/kg) using a 650 kHz transducer (focal spot 3.5 x 22 mm). of AV-45 PET imaging can serve as a diagnostic tool for in vivo quan- Subsequently the FUS set-up was replaced with a set-up for 13C-im- titation of plaque clearance, while ex vivo autoradiography (ARG) can aging, containing a dedicated 13C surface coil inside a small 1H volume provide radio-labelled tracer detectability. This study improves our coil for anatomical reference imaging. An interesting metabolite to understanding of how ultrasound can be used to enhance AD thera- study is pyruvate, which is the product of the glycolysis of glucose and peutic molecule delivery, and promote advances in discovering new is further metabolized in the Krebs’s cycle or converted tolactate. There- therapeutic strategies for neurodegenerative disease. fore, [1-13C] pyruvate was polarized in an in-house-built polarizer as de- scribed previously [Breukels, 2015]. The polarized substrate was rapidly dissolvedin a 1xPBS buffer solution to a final concentration of 80mM. 300μl HP pyruvate was injected within 15s after dissolution. To enable dynamic imaging of the conversion of pyruvate to lactate, a slice select- ive gradient echo sequence was developed that makes it possible to simultaneously image pyruvate and lactate based on their chemical shift difference [Steinseifer 2013]. Every 1.6s an image is acquired that consists of a pyruvate and lactate image [Example, Fig. 1]. A phantom containing [1-13C] pyruvate and [1-13C] lactate was placed next to the animal to obtain a 13C reference image for image registration. From these images, concentration curves of pyruvate and lactate can be ob- tained that by plotting the total signal intensity over time of a region of interest. These concentration curves show the conversion of pyruvate to lactate (Example, Fig. 2). Since gadolinium-based contrast agents Fig. 1 (abstract P7). (A) The conceptual schematics of this study. (B) cannot cross the intact BBB, Magnevist (0.25 ml/kg) was injected after ThioflavinS staining and quantification of amyloidbeta plaque sizes 13C-imaging to verify opening of the BBB using pre- andpost-contrast in the hippocampus area. There was a7.83% (FUS alone), 14.68% (AR T1-weighted MR imaging. alone) and 26.72% (AR+FUS) decrease in comparison to the control RESULTS Pre- and post-contrast T1-weighted images in Fig. 3 confirm group. AR = ARA014418; FUS = focused ultrasound (Fig. 1b) a successful opening of the BBB, as we observed a hyperintense re- gion were the brain was sonicated. We successfully obtained pyru- vate signal in the 13C images, but the intensity of the lactate signals was too low to observe. The pyruvate image did not show any evi- P8 dent signal intensities that can be related to BBB-opening. Novel approach to study metabolic changes after FUS-mediated CONCLUSIONS In this study we successfully showed the feasibility of blood-brain barrier disruption combining BBB disruption using FUS with dynamic imaging of hyper- Thiele Kobus, Tom Peeters, Andor Veltien, Arend Heerschap, polarized 13C-labeled compounds. The BBB was disrupted prior to, and Tom Scheenen remained open during our hyperpolarized 13C-imaging experiment, as Radboud university medical center, Nijmegen, Netherlands confirmed by contrast-enhanced MRI. However,SNR of the hyperpolar- Correspondence: Thiele Kobus ized signals is currently too low to draw conclusions upon enhanced Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P8 uptake or release of hyperpolarized metabolites in the targeted brain region.Therefore, improvement of polarization levels and further OBJECTIVES Focused ultrasound (FUS) in combination with micro- optimization of the imaging parameters is necessary.Despite challenges bubbles can be used to temporarily and locally disrupt the blood- that have to be overcome, this approach enables delivery of hyperpo- brain barrier (BBB) [Hynynen, 2001]. However, little is known about larized agents to the brain and will allow us to study metabolic alter- the exact mechanism of this technique and its effects on the brain. ations due todisruption of the BBB in vivo in the near future. To get more insight, we would like to study metabolic changes after disruption of the BBB. Many metabolites contain carbon-atoms of which approximately 1% is the isotope 13C that can be detected by magnetic resonance imaging (MRI). Due to the low abundance and low gyromagnetic ratio, the sensitivity of 13C MRI is low. Recently, a method has become available to increase the sensitivity of13C-la- beled substrates more than 10.000 fold by dynamic nuclear hyperpo- larization (DNP) and create a solution that can be injected [Ardenkjaer-Larsen, 2003]. After injection of the sample, the conver- sion of the substrate into other metabolites can be observed with 13C MRI enabling the study of fast dynamic metabolic processes in vivo. In this proof of concept, we demonstrate the feasibility of combining FUS-mediated disruption of the BBB with hyperpolarized 13C MRI using DNP. In the near future, this approach might reveal the influence of the BBB on the uptake of hyperpolarized agents or alterations in metabolism due to FUS-mediated BBB opening. METHODS The experimental protocol was approved by the National Animal Research Authority. In two mice, the BBB was disrupted using FUS and followed by hyperpolarized 13C MRI. All experiments took place on a 7T animal system (Clinscan, Bruker). For BBB disruption, animals were anesthetized and positioned in a dedicated MR-guided FUS set-up (IGT, France). MR reference images were acquired for Fig. 1 (abstract P8). Example of a hyperpolarized pyruvate and ultrasound targeting in the left frontal lobe of the brain. The sonic- lactate image mapped to the T2-weighted reference image (no ation (duration=120s, 10ms bursts, burst frequency=1Hz) was per- BBB disruption) formed immediately after the injection of microbubbles (Sonovue, Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 62 of 122 presents control kernel of the system, multiple channels power amplifier, high speed power sensing circuits, low pass filters, and a module of matching circuits. The FPGA-based digital board has three functions. Firstly, it is used as a communication unit with PC through an USB3.0 interface (CYUSB3014, Cypress, USA). Sec- ondly, it can monitor the excitation power by incorporating high speed power sensing module. Lastly, the most important function for the board is to generate multiple channel source waveform and each channel waveform can be controlled individually. The phase information of the waveform is flexible to change so that the system can implement different beamforming algorithms. The power amplifier board is used to amplify the low voltage wave- form to drive the transducer. The low pass filter is connected with the power amplifier board to shape the high-voltage wave- form to match with transducer. A matching circuit is employed to adjust the impedance characteristics to achieve maximal power transfer. And a dynamic phased control beamforming al- gorithm is developed to achieve flexible beam focusing in the acoustic field. RESULTS The prototype of the fabricated ultrasound system is shown Fig. 2 (abstract P8). Example of a dynamic time evaluation of in Fig. 1b. Each module is achieved by separate printed circuit boards hyperpolarized pyruvate and lactate signals (no BBB disruption). The (PCBs). The phase difference error for all channels is less than 7 ns total signal intensity of a region of interest isplotted over time. Data when the center frequency of transducer is set to 800 KHz. It means were not corrected for T1decay and signal loss due to RF pulsing that the phase error of the proposed ultrasound systemis about 2 de- grees, which is able to support a good beamforming performance. Dynamic focusing can be achieved in a circle with a radius of 10 mm, and the focal depth can be adjusted in the range of 4-10 cm. The system can work in a pulsed mode or a continuous mode for 50- 300 ms. The system is able to hit a water column which theheight is about 5 cm. The beamforming quality of the system is also evaluated by an acoustic hydrophone. CONCLUSIONS In this study, we present a prototype design of a low cost ultrasound array system specifically for neuromodulation. Test re- sults demonstrate that the proposed system has a great potential to Fig. 3 (abstract P8). Pre (A) and postcontrast (B) images after gadolinium improve the level of neuromodulation. Future work such as improve- administration. The subtraction image (C) shows a hyperintense region ment of beamforming performance in different acoustic mediums and where the BBB was successfully disrupted. The corresponding 13C image (D) validation of primate animal studies in vivo will be carried out. of hyperpolarized pyruvate after BBB disruption P9 A low-cost phased array system for ultrasound neuromodulation Wu Sun, Juan Zhou, WenBin Yan, JiaXing Yang, Weibao Qiu, Hairong Zheng Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences, Shenzhen, China Correspondence: Wu Sun Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P9 OBJECTIVES Ultrasound has been shown to non-invasively stimulate Fig. 1 (abstract P9). (a) The schematic of the ultrasound electronic and inhibit neuronal activities under various conditions like stimulat- system. (b) the prototype of the proposed ultrasound system ing auditory nerve responses and suppressing sensory-evoked poten- tials in the primary visual cortex. Currently most of the published works for ultrasound neuromodulation are based on single element focused transducer. In contrast with single element transducer, P10 phased array transducer is more flexible to change the target loca- Test study for the electro-acoustic conversion efficiency of focused tion, and potentially to simulate multiple positions simultaneously. ultrasonic transducer based on virtual instrument 1 2, 1 1, 3 1 High intensity focused ultrasound (HIFU) array system can be applied Yang Liu , Jianwen Tan , Deping Zeng , Zhiming Zhong to neuromodulation by decreased the ultrasound energy, however, Biomedical Engineering, Chingqing Medical University, Chongqing, such a system is usually quite expensive and bulky. Thus a low cost China; Chongqing Communication Institute., Chongqing, China; ultrasound array system specifically for ultrasound neuromodulation National Engineering ResearchCenter of Ultrasound Medicine, is needed. This paper presents an ultrasound phased array system Chongqing, China that incorporates a 256-element hemispheric transducer for neuro- Correspondence: Yang Liu modulation. Test results show that the system has potential to Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P10 achieve multiple position neuromodulation. METHODS Our system includes two parts: 1) ultrasound electronic OBJECTIVES The electro-acoustic conversion characteristic of ultrasonic system; 2) phased array transducer (256-element array transducer, transducer is required to be tested when developing focused ultrasound Imasonic, France). The schematic of the ultrasound electronic sys- treatment equipment, with the actual electro-acoustic characteristics of tem is shown in Fig. 1a. The system consists of five major circuit transducer to determine working frequency and driver parameters of fo- modules: an field programmable gate array (FPGA) (Cyclone cused ultrasound therapy system. Therefore, when the HIFU therapy sys- V5CGXFC7D7F31C8, Altera, USA) based digital board which tem developed, there is a massive workload in the measurement of Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 63 of 122 electro-acoustic conversion characteristics of focus ultrasonic transducer. It has put forward higher request to the efficiency and convenience of the electro-acoustic conversion characteristics test system. METHODS An automatic test system for the electro-acoustic conversion characteristic of the transducer based on virtual instrument has been de- veloped in this letter. The input electric power of the transducer is mea- sured by electric power meter based on directional coupler, radiation force balance is used to measure the output soundpower of the trans- ducer. By use of the graphic development environment LabVIEW, an upper computer automatic test software was developed based on virtual instrument technology, which can operate with the real-time data acqui- sition and data processing, analysis and calculation of the data. RESULTS Based on this test system, the electro-acoustic conversion efficiency of the focused ultrasound transducer at different frequencies and the driving power also tested in this paper, the bandwidth of the transducer was ana- lyzed from the point of view that the measured electro-acoustic conversion efficiency of transducer, which discussed electro-acoustic conversion efficiency of transducer with the tracking problem of different driving power and oper- ating frequency based on electrical reflectioncoefficient (Figs. 1, 2, 3). CONCLUSIONS With the analysis of test results shows that the proposed Fig. 3 (abstract P10). The reflectance of transducer under different real-time monitoring tool based on electrical reflection coefficient which driving electric power measured by directional coupler can also applied to tracking the fre- quency of the ultrasonic transducer, which can improve the electrical power transmission efficiency and the output acoustic power of ultra- P11 sonic transducer, and there is important significance on the focused ultra- Multifunctional pulse generator unit for high-intensity focused sound transducer to ensure safety, stability and efficient operation. ultrasound system 2,1 2 2 Satoshi Tamano , Shin Yoshizawa , Shin-ichiro Umemura 1 2 Hitachi, Ltd., Kokubunji, Japan; Tohoku Univ., Sendai, Japan Correspondence: Satoshi Tamano Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P11 OBJECTIVES High-intensity focused ultrasound (HIFU) therapeutic systems have made it possible to treat diseases with fine spatial reso- lution. One critical technical challenge of HIFU is to generate high- voltage, large-current, multi-channel pulsed signals to effectively ex- cite the low-impedance HIFU transducers. This paper presents the development of multifunctional pulse generator unit for HIFU system. The pulse generator can produce pseudo sinusoidal wave, triggered- HIFU mode pulse and second-harmonics superimposed shock wave lithotripsy pulses for HIFU transmission. We report the development of a novel multifunctional, reconfigurable pulse generator using FPGA and high-voltage MOSFETs. METHODS The pulse generator is interfaced with a PC and receives commands, waveform data, focusing delay data through a high-speed USB 2.0 micro controller. The USB microcontroller transmits communi- cation information with the PC to the control FPGA. The control FPGA Fig. 1 (abstract P10). The electro-acoustic conversion efficiency and stores instruction information from the PC internalregisters, and de- reflectance of focused ultrasound transducer at different frequencies livers transmission waveform information and transmission focus data to the transmission FPGA. For data communication between the con- trol FPGA andthe transmission FPGAs, information is transferred using a dedicated local bus. In our system, 32 channels transmission can be performed per transmission board. Sincewe use HIFU transducer con- sisting of 128 elements, we used four transmission boards. The control board performs generation of actual ultrasonic wave sequences and ar- bitration of the transmission board control bus as well as the functions described above. The transmission board includes a transmission FPGA for generating a timing signal for ultrasonic transmission based on a command transmitted from the control FPGA, RAMs that stores trans- mission focus data, N-channel and P-channelMOSFETs that actually generates 400 Vpp ultrasonic transmission signal, MOSFET drivers for driving MOSFET. In this study, the operation of the transmission circuit- was evaluated using our prototype dual transmission frequencies trans- ducer. This prototype transducer consists of seven elements that can transmit one and two MHz ultrasonic waves. With these two transmis- sion frequencies, since the impedance of the transducer is around 150 Ω, it is necessary to apply a transmission circuit withlarge driving capability. RESULTS In pseudo sinusoidal transmission waveform mode, the test results show that the power consumption is reduced by 14.8 % and Fig. 2 (abstract P10). The electro-acoustic conversion efficiency and MOSFET maximum temperature rise is reduced by 11.5 °C by using reflectance of transducer under different driving electric power the newly proposed circuit than the our previous class D circuit. Also, Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 64 of 122 it can be seen that the third and fifth harmonic components can be P13 suppressed by 23.9, 30.5 dB, respectively. Therefore, the transmission Axial controllable multiple traps of acoustic vortices generated by waveform becomes closer to the sinusoidal waveform. As a result of directional sources 1 1 1 2 2 the above, we quantitatively evaluated the power saving effect by in- Yuzhi Li , Gepu Guo , Qingyu Ma , Juan Tu , Dong Zhang 1 2 creasing the number of power supply steps and the effect of reducing Nanjing Normal University, Nanjing, China; Nanjing University, Nanjing, the device temperature rise. In the triggered-HIFU mode, high power China (>300 Vpp) and a short time (several microseconds) ultrasonic radiation Correspondence: Yuzhi Li called trigger-burst, and a medium output (< 100 Vpp) andlong dur- Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P13 ation (several seconds) ultrasonic radiation called heating-waves are used in combination. In the transmission circuit, heating-waves has a OBJECTIVES Characterized by the pressure circle and phase spiral, larger amount of heat generation than trigger-burst, so there was a big acoustic vortex (AV) can be used to manipulate objects with its or- problem how to realize a circuit with reduced device power con- bital angular momentum androtation torque. Compared with light, sumption. The proposed circuit can 18.9% lower power consumption acoustic wave can go into media non-destructively with deeper than conventional class D circuit, and 33.6 °C suppress MOSFET penetration depth, which makes it possible to manipulate particles temperature rise. In peak negative enhanced second-harmonics super- inside object using AV, exhibiting a prosperous future in biomedical imposed shock wave lithotripsy mode, by using the proposed circuit, it engineering. In previous studies, AV was often investigated under was possible to realize a concavity of the positive voltage side output, the framework of point source radiation with acoustic diffraction. and the second harmonic ratio approached thetheoretical value 2.0 dB However, the point source based model is not practical for direc- from the class D circuit. tional sources with big ka values, which are influenced by beam- CONCLUSIONS We modified the HIFU ultrasonic transmission circuit patterns with obvious side lobes. Consequently, more attentions previously reported at ISTU 2016 and created a HIFU transmission should be focused on the distributions of AV generated by direc- unit with 128 channelsintegrated. In particular, by increasing the tional sources. transmission voltage level from four levels to seven levels, MOSFET METHODS The phase coded approach is employed to generate con- device heat generation could be greatly reduced. At the same time, trollable AV using directional sources. Based on the radiation theory we achieved a significant reduction in power consumption. As a re- of planar piston source andcoded initial phase, the physical mechan- sult, the risk of MOSFET damage is reduced, and HIFU transmission ism of AV is theoretically investigated with explicit formulae. The could be executedstably and safety. In our newly proposed circuit, it principles of main-AV (M-AV) and vice-AVs (V-AVs) generated by the was demonstrated that it is effective not only for transmission of main lobes and the side lobs of the sources are analyzed. The num- pseudo sinusoidal wave but also for triggered-HIFUmode and sec- ber and locations of the formed M-AV, V-AVs and the corresponding ond-harmonics superimposed shock wave lithotripsy. The applicabil- vortex valleys (VVs) are calculated for different ka values. And the ity of our proposed circuit has expanded. generations of axially controllable multiple traps are discussed based on Gorkov’s theory. The proposed theory isalso verified by numerical studies and experimental measurements for different ka values at P12 the frequency of 1 MHz. The improvement effect of magnetic microbubbles on HIFU- RESULTS It is proved that obvious M-AV, V-AV and VVs can be gener- induced hyperthermia effect ated for higher ka with the least value of 3.83. Corresponding to the Dongxin Yang, Yanye Yang, Guangyao Xu, Xiasheng Guo, Juan Tu, Dong 8-source experimental systemfor ka=29.32, the measured axial pro- Zhang files and the cross-sectional distributions at different heights show Nanjing University, Nanjing, China good agreements with the simulated ones, and obvious phase spirals Correspondence: Dongxin Yang of M-AV and V-AVs are also demonstrated. Several VVs with almost Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P12 pressure zero are also observed on the center axis to form multiple traps, which can beaxially controlled for ka adjustment. OBJECTIVES Encapsulated microbubbles (MBs) have been widely CONCLUSIONS It is demonstrated that, for bigger ka, besides M-AV used to enhance high intensity focused ultrasound (HIFU) -induced generated by the main lobes of the sources, cone-shaped V-AVs pro- hyperthermia by making use of its ability of increasing acoustic en- duced by the side lobes are closer to the source plane at relatively ergy absorption and lowering the cavitation threshold. To balance lower pressure. Corresponding to the radiation angles of press valleys the needs of treatment efficiency and safety, there is increasing de- between the main lobe and the side lobes of sources, VVs with al- mand of more efficient encapsulated MB agents that can quickly most pressure zero can be generated on the central axis to form ax- achieve sufficient hyperthermia effect while minimizing the damage ially controllable multiple traps with the locations controlled by ka to surrounding tissues. adjustment. The results provide the feasibility of deep-level control METHODS In the present work, superparamagnetic iron oxide nano- of AV inside object and suggest the application potential of multiple particles (SPIO) were coupled to perfluorocarbon-filled, albumin-en- traps for particle manipulation in biomedical engineering. capsulated microbubbles (referred as SPIO-albumin MBs) to enable a stronger enhancement of the HIFU-induced hyperthermia effect than regular albumin-encapsulated ones. The thermal enhancement cap- P14 acity of SPIO-albumin MBs and albumin-encapsulated MBs was inves- Study on the acute injury effect on candida albicans by low- tigated based on both experimental measurements and numerical frequency and low-intensity ultrasound simulations ofthe temperature change at HIFU focus. Yang Xiang RESULTS The results show that, comparing with the use of albumin-en- Chongqing Medical University, Chongqing, China capsulated MBs, the adoption of SPIO-albumin MBs will bring about Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P14 quicker temperatureelevation rate and higher peak temperature. CONCLUSIONS An improvement can be made to the enhancement OBJECTIVES To investigate the effects of Low-frequency and Low-inten- of the HIFU-induced hyperthermia by using SPIO-albumin MBs rather sity Ultrasound (LFLIU) on the acute injury of candida albicans, and to in- than albumin-encapsulated ones, which is because the thermal prop- vestigate the effect of LFLIU on the permeability of the cell wall. erties of the two kinds of microbubbles are different. With these ad- METHODS Concentration is 1.5 x 107 cfu/ml of Candida albicans bacteria vantages, these SPIO-albumin MBs can be introduced tospecific liquid, with 5 ml bacteria to single flageolet culture plate. With the fre- locations of interest to intensify the thermal effect of HIFU much quency for 42 kHz, probe diameter is 5 cm circular planar ultrasonic treat- more efficiently, which might enable more ideal non-invasive con- ment head, ultrasonic intensity was selected 0.13 W/cm2, 0.35 W/cm2 trollable hyperthermia treatment strategies and applications. irradiation six well culture plate beads bacterium solution for 5 min, After Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 65 of 122 48 h count ultrasound irradiation on petri dish culture survival of colonies of transducer on MAT-MI and provide the fundamentals for transducer in a petri dish. Transmission electron microscope and scanning electron selection in further study to improve the accuracy of electrical imped- microscope were used to observe the external shape and internal struc- ance reconstruction. ture of the bacteria. RESULTS Different doses of ultrasound irradiation 5 min, 48 h after culture dish colony average count shows, Control group colony P16 count to 21 cfu, ultrasonic intensity0.13 W/cm2 group of colony A method for evaluating the relationship between the inertial count for 20 cfu, ultrasonic intensity of 0.35 W/cm2 group of colony cavitation duration and the acoustic parameters 1 1 2 1 count for 14 cfu.LFLIU candida albicans, scanning electron microsco- Mouwen Cheng , Yutong Lin , Alfred C. Yu , Peng Qin pe(SEM) visible thalli were swollen shape becomes large, Under the Department of Instrument Science and Engineering, Shanghai Jiao scanning electron microscope (SEM) visible thalli was swelling, the Tong University, Shanghai, China; Department of Electrical and shape becomes large, the extracellular fluid into the cells increased Computer Engineering, TheUniversity of waterloo, Waterloo, Alberta, obviously; Transmission electron microscope (TEM) see nuclear mat- Canada ter at the edge of the aggregation, cell membrane damage, forma- Correspondence: Mouwen Cheng tion of vacuoles, large amounts of water into cells. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P16 CONCLUSIONS With the increase of ultrasonic irradiation intensity, signifi- cantly increased the mortality rate of Candida albicans; Candida albicans OBJECTIVE Inertial cavitation, triggered by the ultrasound and micro- in LFIU can promote the increase of the permeability of the cell wall of. bubbles, is considered to be the main mechanism for sonoporation- mediated macromolecule delivery. Inertial cavitation dose in most studies has been employed to evaluate the delivery efficiency and treatment efficiency. However, the temporal characteristic of the iner- P15 tial cavitation are also closely related to the bioeffects accompanied Transducer directivity influence on artifacts reduction for by sonoporation. This study aims to propose a method for evaluating magnetoacoustic tomography with magnetic induction the temporal duration of inertialcavitation and determine its relation 1 1 2 2 Gepu Guo , Qingyu Ma , Juan Tu , Dong Zhang with the acoustic parameters. Nanjing Normal University, School of Physics and Technology, Nanjing, METHODS An agarose-gel tissue-mimicking phantom was fabricated to Jiangsu, China; Nanjing University, Nanjing, China hold 1% SonoVue microbubbles solution. 1-MHz plane transmission Correspondence: Gepu Guo transducer and another 7.5-MHzfocused transducer were employed for Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P15 triggering the cavitation of microbubbles and passively detecting acous- tic signal, respectively. Then the signal was amplified and recorded by a OBJECTIVES As a novel noninvasive modality of detecting electrical high-speed digitizer. After the frequency domains characteristics of the conductivity variation for tissues, magnetoacoustic tomography with signals were analyzed, the distribution of the broadband energies during magnetic induction (MATMI) has been demonstrated to have the cap- the exposure time was obtained (Fig. 1). The full width at the half max- ability of distinguishing the early pathological changes of biological tis- imum in the time trace of broadband energy was proposed to deter- sues inside the object. In previous studies, the transducer was usually mine the temporal duration of the inertial cavitation behavior. simplified as an ideal or omnidirectional receiver without the consider- RESULTS This study determine: 1) Peak rarefactional pressure (PRP) was ation of its directivity and scanning radius. However, the properties of negatively correlated with inertial cavitation duration. Inertial cavitation transducer play a vital role in signal acquisition and image reconstruc- duration rapidly decreased from about 29.31 ms to 1.50 ms for the in- tion. In order to optimize image reconstruction and eliminate the creasing PRP from 0.5 MPa to 0.7 MPa. But for the PRP above 0.7 MPa, in- image artifacts for MAT-MI, the influence of transducer was investigated ertial cavitation duration approximately approaches to 1.26 ms, suggested theoretically for a two-layer eccentric spherical tissue model based on inertial cavitation duration tended to be saturated. 2) Inertial cavitation the principles of acoustic dipole radiation and transducer reception. duration gradually increased from 0.60 ms to 1.23 ms with the increasing METHODS Based on the principles of magnetic excitation, acoustic vi- pulse duration (PD) from 10 μs to 400 μs, and was positively correlated bration, acoustic dipole radiation and transducer reception, numerical with the PD. 3) Pulse repetition frequency (PRF) exhibited relatively weak- simulations are performed for a two-layer eccentric spherical phantom endependent than PD on the inertial cavitation duration (Figs. 2, 3). model. The factors that affect transducer directivity are analyzed, and CONCLUSIONS The proposed inertial cavitation duration could be the distributions of acoustic pressure and waveform are simulated used to evaluate the kinetics of the inertial cavitation triggered by using the transducers with omni-directivity, strong-directivity and uni- pulsed ultrasound and microbubbles. The relationship between directivity. Then the MAT-MI images reconstruct with the collected acoustic parameters (PRP, PD, PRF) and inertial cavitation duration of acoustic waveforms are achieved and compared to the corresponding SonoVue microbubbles were determined. These finding suggested model to analyze the affect of artifacts reduction. inertial cavitation duration, in combination with the previous inertial RESULTS It is demonstrated that the image artifacts of MAT-MI is obvi- cavitation dose, would be the important factors for evaluating cavita- ous for waveform collection using omni-directional transducer. Accord- tion-mediated therapy. ing with the increase of transducer directivity, the detection angle of the receiver decreases with an increased sensitivity. Especially for a uni-direc- tional transducer, the collected pressure reflectsthe strength of acoustic vibration along the normal direction of the receiver, which can be used to reconstruct the conductivity contrast image without artifacts. In prac- tical applications, large-radius transducer with strong directivity can be applied as an approximate uni-directional receiver to improve image quality with little artifacts aspossible. In addition, to realize narrow detec- tion scope with a fixed transducer, the scanning radius should also be optimized to achieve acceptable signal to noise ratioand peak pressure ratio. The favorable results confirm the influence of transducer on MAT- MI and provide the fundamentals for transducer selection in further study to improve the accuracy of electrical impedance reconstruction. CONCLUSIONS It is concluded that large-radius transducer with strong directivity can be applied as an approximate uni-directional receiver to improve image quality with little artifacts as possible. In addition, to realize narrow detection scope with a fixed transducer, the scanning ra- dius should also be optimized to achieve acceptable signal to noise ra- Fig. 1 (abstract P16). See text for description tio and peak pressure ratio. The favorable results confirm the influence Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 66 of 122 P17 by time reversal method using simulation. At first the sound source Sub-wavelength focal region achieved by a spherical focused is set at the target point at the simulation model constructed by ultrasound transducer with open ends in resonator modes monkey’s CT data and the ultrasound emitted and propagates to the 1 1 2 1 3 3 Hua Cao , Min He , Zhou Lin , man luo , Guangrong Lei , Xiaobo Gong , array transducer through the skull. Next the received signals are re- 4 1 1 5 2 Jun Dang , Deping Zeng , Faqi Li , Junru Wu , Dong Zhang , versed and emitted from the array transducer to the target, which Zhibiao Wang focus the ultrasound on the target correctly. Simulation solves the 1 2 Chongqing medical university, Chongqing, China; Institute of basic equation of continuum mechanics by FDTD method. And it an- Acoustics, Key Laboratory of Modern Acoustics, MOE, Nanjing University, alyzes the ultrasound propagation by treating the medium as the Nanjing, China; National Engineering Research Center of Ultrasound mixture of water and bone. In order to model the complicated living Medicine, Chongqing, China; Department of Oncology, 1st Affiliated body tissues Hounsfield unit of CT images is translated to the volume Hospital of Chongqing Medical University, Chongqing, China; Physics, fraction of bone. The density and sound speed of each voxel is calcu- School of Engineering, the University of Vermont, Burlington, Vermont, lated by the volume fraction. It allows to analyze very large scale cal- United States culation rapidly. Correspondence: Hua Cao RESULTS To investigate the effect of phase controlling the simulation Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P17 using actual macaque monkey’s CT data is conducted. Figure 1 is simulation results and each image shows the pressure distribution OBJECTIVES High intensity focused ultrasound (HIFU) has become a surrounding the target point. Intensity of pressure is normalized by new noninvasive surgical modality for cancer treatment, however, maximum value of focal point. (a) shows the case of only waterme- the HIFU focusing precision is handicapped by the diffraction limit of dium and (b), (c) show the result under existence of bone without the wavelength of a traveling ultrasonic wave. A new HIFU trans- control and with control respectively. The focusing with control is im- ducer has been developed, which uses standing waves generated in proved compared to focusing without control. Figure 2 shows the re- a spherical cavity with open ends and break the diffraction limit to sult of 3 different ch number of array transducer. Compared with achieve subwavelength focusing region. 16ch, 64ch has better focusing but there is no large difference be- METHODS In order to describe the acoustic field, a finite element tween 64ch and 128ch. model is developed to numerically study the acoustic field gener- CONCLUSIONS Simulation result suggests that the phase controlling ated from the spherical cavity transducer assembly and the ex- can work effectively. And it is also founded that by comparing the re- periment was setup to measure the frequency dependence of sult of 3 different number ofch, 64ch is enough to control the focal the acoustic field generated by this spherical cavity transducer. point. In the future in order to verify the simulation result we are go- RESULTS Our theoretical and experimental results demonstrate ing to compare it with experimental measurement. Then parameters that in its resonant modes, the focusing zone is smaller while the and modeling method of simulation are reconsidered for the experi- focusing gain of sound pressure ishigher. ment using macaque monkey. CONCLUSIONS These results indicate that the focal zone of the acoustic field inside a spherical cavity with open ends is compressed to a sub-wavelength level while the intensity of sound pressure in the focal region significantly increases. P18 Non-invasive therapeutic method for brain disease using TFUS system assisted by numerical simulation 1 2, 1 3 3 Yohei Kobayashi , Takashi Azuma , Tatsuya Umeda , Tomomichi Oya , 3 4 1 4 Masashi Koizumi , Ryo Suzuki , Naoto Yamamura , Kazuo Maruyama , 3 1 Kazuhiko Seki , Shu Takagi 1 2 Bioengineering Dept., The University of Tokyo, Tokyo, Japan; Faculty of Medicine, The University of Tokyo, Tokyo, Japan; National Center of Neurology andPsychiatry, Kodiara-shi, Tokyo, Japan; Teikyo University, Tokyo, Japan Correspondence: Yohei Kobayashi Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P18 OBJECTIVES tFUS (transcranial focused ultrasound) has a great po- tential for non-invasive therapy for brain disease. The purpose of this research is to apply tFUS with microbubble for BBB opening for DDS and stimulation of red nucleus for motion trigger. BBB opening is the technique to enhance the permeability of blood brain barrier for effi- cient drug delivery. On the other hand, stimulation of red nucleus which exists in deep area of brain is said to be effective against re- habilitation from cerebralinfarction. For investigating these therapy, we conduct the experiment using macaque monkey which is closer to the human. In either case major issues concerning application of tFUS for brain therapy is focal displacement due to reflection and re- fraction through the skull. In this research we develop the focal con- trolling method with array transducer and simulation of ultrasound propagation utilizing CT data of macaque monkey. Fig. 1 (abstract P18). Pressure distribution surrounding target METHODS Array transducer can control the focal point by adjusting point(a) Water (b) skull without control (c) skull with control phase of each element. Phase delay of each element can be decided Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 67 of 122 harmonic responses grow with increasing acoustic pressure for both chirp and sinusoidal excitations and this growth would reach a satur- ation level when acoustic pressure exceeds a specific threshold.The ex- perimental results agree with simulations when the driving pressure is less than 300 kPa, but the saturation amplitudes of sub-harmonic re- sponses for both excitations after that are much lower than the simu- lated results. When the driving pressure is greater than 100 kPa and less than 300 kPa, the sub-harmonic amplitudes excited by chirp signals are always 5-10 dB higher than those excited by sinusoidal signals. The stable cavitation threshold for chirp excitation is also observed to be much lower than for sinusoidal excitation. Optimization: As the band- width of chirp signal increase, the sub-harmonic amplitudes changes more greatly when the ambient pressure changes. This indicates that the wider bandwidth could offer a better sensitivity in ambient pres- sure evaluation by exciting a wider size range of microbubbles. How- ever, bandwidths over 22.8% were not investigated because of the overlap effect wider bandwidths may bring. The studies on chirp pulse length indicates that the measured sub-harmonic responses increase with longer pulse length. The inherent reason for this behavior is that each individual microbubble may take a few cycles to reach a steady state when a strong non-linear effect is observed. CONCLUSIONS The sub-harmonic waves produced by stable cavita- tion of bubbles provides the feasibility of enhanced blood pressure measurement. Due the non-negligible size distribution of commercial ultrasound contrast agent microbubbles, chirp signals rather than si- nusoidal signals are applied in this article to excite a wider size range of microbubbles. Chirp ultrasound has been theoretically and experi- mentally verified to effectively enhance both the amplitude of sub- Fig. 2 (abstract P18). Pressure distribution surrounding target harmonic response and its sensitivity to ambient pressure, thus it pointof three different number of ch overweighs conventional sinusoidal signals in pressure measurement. Wider bandwidths and longer pulse lengths for chirp excitation also prove to realize an optimized pressure evaluation routine. Further P19 potential development of the proposed method will require en- Evaluation of ambient pressure using sub-harmonic response of hancement of bubble dynamics models and improved microbubble microbubbles sonicated with chirp pulses fabrication techniques so that this method could work better for Zhiyang Jin, Siyu Liu, Xiasheng Guo, Juan Tu, Dong Zhang non-invasive blood pressure measurement applications. Nanjing University, Nanjing, China Correspondence: Zhiyang Jin Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P19 OBJECTIVES The sub-harmonic signal generated by ultrasound contrast agents’ stable cavitation has been proven to be a potentially effective and efficient cue for noninvasive blood pressure measurement. In this work, an improved ambient pressure evaluation method is proposed to enhance the sub-harmonic responses of microbubbles so that a more sensitive and accurate pressure measurement could be achieved. METHODS Chirp signals, namely, linear frequency-modulated signals, are combined with microbubble sub-harmonics in this work. Both simulations and experiments are conducted to illustrate the advan- tage of chirp excitation in ambient pressure evaluation by compari- sons with conventional sinusoidal excitation. Dependence of subharmonic response on chirp parameters, namely, acoustic pres- sure, central frequency, bandwidth and pulse length, are also studied for optimization of sub-harmonic responses under chirp sonication. All the simulations are based on Marmottant model, supposing microbubbles have a Gaussian size distribution. Commercially availa- Fig. 1 (abstract P19). See text for description bleSonoVue microbubbles (Bracco Diagnostics, Geneva, Switzerland) were used for the experimental measurements and the experimental setup is illustrated in Fig. 1 uploaded. RESULTS Sub-harmonic enhancement by chirp excitation: SonoVue microbubbles are driven by sinusoidal and chirp excitation sharing the P20 same driving pressureamplitude and pulse length at Pov = 0 kPa (spe- Development of a focused ultrasound device for skin ablation 2 1 2 1 cifically, ambient pressure of 1 atm). The sinusoidal wave frequency Meng-Hung Tsai , Li-Chen Chiu , Win-Li Lin , Gin-Shin Chen and the central frequency of the chirp excitation werefs= fc = 3.5 MHz, Institute of Biomedical Engineering and Nanomedicine, National Health and the chirp signal bandwidth was Δf = 0.4 MHz. A significant increase Research Institutes; Institute of Biomedical Engineering, NationalTaiwan of 5.1 dB in the sub-harmonic component can be observed in the chirp University excitation case. As the ambient pressure increases, the measured sub- Correspondence: Meng-Hung Tsai harmonic responses induced by chirp excitation reduce more violently Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P20 than sinusoidal case, providing better sensitivity in ambient pressure measurement. All the experimental results agree well with the OBJECTIVES Focused ultrasound can concentrate acoustic power on simulations. Dependence of sub-harmonic response on driving acoustic the target tissues like subcutaneous fat and superficial muscular apo- pressure: In a similar setup with the former one, the amplitudes of sub- neurotic system to generate a steep temperature elevation, leading to Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 68 of 122 collagen denaturation, contraction and remodeling. In the study, a sin- gle-element focused ultrasound transducer integrated with an auto- matic motion system was developed for multi-point ablation in the skin. METHODS The transducer was made of 1-3 piezocomposite mate- rials with a diameter of 10 mm and a radius of curvature of 17 mm (Fig. 1). The operation frequency was 5MHz. The electro- acoustic conversion efficiency and focal zone were measured after electrical impedance matching. The effectiveness of the de- vice was evaluated by the ablation experiments of phantom, ex- vivo and in vivo rat model. RESULTS The efficiency of the transducer was 43.9±2.6%. The focal depth and focal width were 7.2 and 0.7 mm, respectively. With the ultrasonic parameters of electrical power of 10 W for 1 s, the lesions were formed in the 52°C thermo-sensitive hydrogel phantom (Fig. 2). Fig. 3 (abstract P20). The H&E staining of the rat skin tissue after Ex vivo and in vivo studies showed that the transducer could pro- ultrasonic sonications duce a hot spot to noninvasively cause superficial skin necrosis as the electrical power and time of ultrasonic sonication were in a range of 6-9 W and 2 s. CONCLUSIONS A relatively high-frequency focused ultrasound device P21 has been developed for skin ablation. In vitro and animal studies Low intensity pulsed ultrasound stimulates hair follicle cells in 3D have verified the efficacy ofthe ultrasonic device (Fig. 3). culture Noboru Sasaki, Mitsuyoshi Takiguchi Hokkaido University, Sapporo, Japan Correspondence: Noboru Sasaki Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P21 OBJECTIVES Low intensity pulsed ultrasound (LIPUS) has been known to activate protein synthesis, DNA synthesis and cell prolifera- tion in several different types of cells. This proof of concept study evaluated whether LIPUS stimulates proliferation of hair follicle cells in 3D culture. METHODS The 3D culture was composed of three layers. In the top layer, Human Follicle Dermal Papilla cells (HFDPC; PromoCell) were embedded into Matrigel (Corning). The middle layer consisted of only Matrigel. In the bottom layer, Human Hair Outer Root Sheath cells (HHORSC; ScienCell) were embedded into Matigel.Low intensity pulsed ultrasound was exposed to cells from above the 3D culture. Ultrasound parameters are as follows; 1 MHz center frequency, 500 pulses, 1 kHz PRF (i.e. the duty factor was 50%), Isata 90 mW/cm2. After ultrasound exposure, cells in the 3D culture were stained by Calcein-AM and observed by a fluorescent microscopy. RESULTS Both HFDPC and HHORSC were increased by single expos- ure of LIPUS. Moreover, HFDPC grew upward, i.e. from the bottom Fig. 1 (abstract P20). The photo of the 5MHz focused layer to the top layer. ultrasonic transducer CONCLUSIONS This study showed the feasibility of LIPUS for stimu- lating the proliferation of hair follicle cells. Further in depth study may assess mechanisms of thestimulation and optimize ultrasound parameters. P22 MR thermometry of a novel focused ultrasound application in post-mortem skin 1 2 Ziqi Wu , Luis Hernandez-Garcia 1 2 Access Business Group, Grand Rapids, Michigan, USA; Biomedical Engineering, University of Michigan, Ann Arbor, Michigan, USA Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P22 OBJECTIVES Intense focused ultrasound has been introduced in past years for non-invasive anti-aging facial treatment. By creating micro- necrosis in dermalor superficial muscular aponeurotic system (SMAS), wound healing process can be triggered followed by promotion of fi- broblasts activity and collagenproduction. Recent research showed that repeated mild application of focused ultrasound energy in the skin with lower acoustic intensity also provided clinical benefits. Temperature rise generated inside the skin using this technique is hypothesized as the mechanism of action, and the estimation of temperature within the targeted region is important for the applica- Fig. 2 (abstract P20). A white lesion was induced in the tion efficacy and safety. In this study, we demonstrated that MR thermalsensitive hydrogel phantom by the developed transducer thermometry can be used to measure the 2Dtemperature changes Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 69 of 122 generated by a novel focused ultrasound skin applicator with fine spatiotemporal resolutions in post-mortem skin. METHODS A MR compatible prototype transducer was constructed (Ac- cess Business Group, USA) using a partial cylindrical ceramic element and a plastic waveguide. The focal distance was confirmed to be 4 mm in water by Schlieren imaging and the -6dB focal region was measured to be 1mm inwidth and20mmin lengthusing ascanninghydrophone system. The maximum rarefraction pressure was less than 1 MPa while the spatial-peak pulse-average intensity was less than 30 W/cm2. The transducer was driven with a function generator and an amplifier at a center frequency of 4.5 MHz with 2.8 W average acoustic power. Pulsed ultrasound (400 ms) was generated with a duty cycle of 89% and total Fig. 2 (abstract P22). (A) MR image acquired using gems sequence. exposure time varied from 7 to 30 seconds. A 7T MR scanner (Agilent- Postmortem full thickness skin is roughly 1 cm thick overlaid on top Technologies, Walnut Creek, CA) was used to image porcine samples for of a piece of porcine muscle which helpedavoiding reflection from repeatability of MR thermometry and followed by imaging the post-mor- the bottom of the chamber. (B) Trace of the maximum temperature tem skin. Highresolution MR images were acquired with multislice gradi- rise in the skin from a 30 second ultrasound exposure ent-echo sequence (TR/TE = 20/4 ms, flip angle = 20, voxel size = 0.47 x 0.47 x 2 mm) at the beginning and end of the experiment while gradient echo EPI sequence (TR/TE = 400/11.32 ms, triple shots, voxel size = 0.42 x 0.83 x 1 mm) were performed before, during, and after ultrasound expos- P23 ure. The ultrasound transducer and the MRI scanner were synchronized Ultrasound-mediated transdermal delivery of hyaluronic acid into via TTL pulses, such that the EPI images were collected between ultra- skin sound bursts. Spatiotemporal temperature changes were computed Yi Yun using proton resonant frequency shift relationship from the MR phase VITA-Sound Tech, USA images. Finally, skin samples were stored in 10% formalin, fixed in paraf- Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P23 fin, sliced, and stained with H&E and masson’s trichrome to investigate thermal damage on skin cells. Transdermal drug delivery (TDD) can effectively bypass the first-pass RESULTS The reconstructed temperature images in the coronal plane effect, which can be valuable in cosmetic industry. In our work, across the ultrasound beam width showed the best image quality (Fig. 1) ultrasound-facilitated TDD on fresh porcine skin was studied in vari- whereas the sagittal plane images along the beam length were poor due ous conditions of acoustic parameters. The delivery of fluorescent to magnetic field inhomogeneity. In the porcine muscle, maximum nanoparticles and high molecular weight hyaluronic acid (HA) in the temperature rise of 13.2 ± 0.3 °Coccurred inside the focal region and the skin samples was observed by laser confocal microscopy and ultra- MR thermometry results were repeatable with four subsequent ultra- violet spectrometry, respectively. The results showed that, with the sound exposures. Transient changes were also observed on MR magni- application of ultrasound exposures, the permeability of the skin to tude images. In the post mortem skin, 8 seconds ultrasound exposure these markers (e.g., their penetration depth and concentration) could generated temperature rise of 17.9 °C whereas 30 seconds exposure be obviously raised above its passive diffusion permeability. More- caused 33.3 °C temperature increase (Fig. 2). The maximum temperature over, ultrasound-facilitated TDD was also tested with/without the rise occurred at 2 mm from the transducer surface into the dermis. No presence of ultrasound contrast agents (UCAs). When the ultrasound permanent histological cell damage was seen for shorter ultrasound ex- was applied without UCAs, low ultrasound frequency will give better posure whereas necrosis was observed with longer ultrasound exposure. drug delivery effect than high frequency, but the penetration depth CONCLUSIONS Using gradient echo EPI sequence, MR can measure was in a less level around 200 μm. However, with the help of the spatiotemporal temperature changes induced by a novel ultra- ultrasound-induced microbubble cavitation effect, both the penetra- sound skin applicator inpost-mortem skin and provide potential tion depth and concentration in the skin were significantly enhanced safety temperature or thermal dose threshold for focused ultrasound even more. The best ultrasound-facilitated TDD could be achieved dermatological applications. (A) Reconstructed temperature maps with a drug penetration depth of over 600 μm, and the penetration based on PRF shift superimposed with high resolution gem image; concentrations of fluorescent nanoparticles and HA increased up to (B) Traces of the maximum temperature. about 4-5 folds. In order to get better understanding of ultrasound- facilitated TDD, scanning electron microscopy was used to examine the surface morphology of skin samples, which showed that the skin structure changed greatly under the treatment of ultrasound and UCA. The present work suggests that, for TDD applications (e.g., nanoparticle drug carriers, transdermal patches and cosmetics) in cosmetic industry, protocols and methods presented had shown us the potentially attractive application for moisture and treatment of Skin Deapth. P24 Comparative study on ultrasonic monitoring of pHIFU and cHIFU peripheral ablation mode Wen Jing Chongqing Medical University, Chongqing, China Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P24 Fig. 1 (abstract P22). (A) Reconstructed temperature maps based OBJECTIVES To investigate the feasibility of Ultrasonic monitoring on on PRF shift superimposed with high resolution gem image; (B) PHIFU and CHIFU peripheral ablation mode. Traces of the maximum temperature rise measured in porcine METHODS 60 cases ox-liver tissues were divided into group A muscle from 4 subsequent ultrasound exposures (PHIFU, DC=15%, n=30) and group B(CHIFU,DC=100%,n=30).Under Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 70 of 122 the guidance of B mode ultrasound, each group was performed with P26 peripheral ablation, during and after the ablation, capture images of Concentration of MSNC-PFP influence surface models in high-intensity every single line and layer related to the target tissue, and then focused ultrasound ablation of ex vitro bovine liver analyze the gray scale change. After the whole tissue peripheral abla- Ding Xiaoya, Dazhao Ma, Wen Jing, Qi Wang, Jianzhong Zou tion, slice the target and observe its damage situation. Chongqing medical university, Chongqing, China RESULTS In the course of peripheral ablation, both groups were seen in- Correspondence: Ding Xiaoya stant echo enhancement in the irradiated areas, with time varying, the Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P26 Hyperecho were finally replaced by hypoecho in group A and the tissues were presented liquefaction necrosis, while linear Hyperecho were still OBJECTIVES To Discusse the influence of perfluoropentane drops-en- found in group B and tissues were presented coagulation necrosis. After capsulated mesoporous silica nanocapsules (MSNc-PFP) injected into peripheral ablation, a cyclic hypo-echo were observed around the target the melting region withthe surface ablation of HIFU in vitro bovine and the gray scale of the internally areas were slightly changed and tis- liver. sue observing nearly showed damaged. While in group B, the instant METHODS Ox-liver tissues were randomly divided into 5 groups: PBS Ultrasonic monitoring images were not agreed with the form of the tis- blank control group, 0.25 mg/ml group, 0.5 mg/ml group, 1 mg/ml sue damage situation for the entire target were presented time group, 2 mg/ml group according to the concentration of MSNc-PFP, depended strong echo change, and the areas where there is actually co- melting line isometric injection five points, measured 100 uL. Each agulation necrosis were partial hypoecho or medium echo. group was surface ablationed that the edge internal were not in- CONCLUSIONS The tissue damage character, ultrasonogram and gray cluded with the same dose of HIFU energy after rejection. scale change were all varied between PHIFU and CHIFU peripheral RESULTS 1 mg/ml group, 2 mg/ml group, the high-level echo of syn- ablation mode. It is possible to apply Ultrasonic monitoring while ergistic agent in the injection area was receding after injection; After performing peripheral ablation mode, but there is still limitation. HIFU irradiation, regional gray values were heighten and the scope was broadening changing with the concentration of synergist in- creased. Each group can form a complete coagulation necrosisexcept P25 the control group. And the higher the synergistic agent concentra- Experimental study of the effect of the target blood vessels angled tion used, the width and the ablation range of oagulation necrosis with acoustic axis on the surface ablation of pHIFU were greater. 2 mg/ml group showed obvious thermal damage per- Yang S. Ying, Zou Jianzhong ipheral ablation region. Imaging and Nuclear Medicine, Institute of biomedical engineering, CONCLUSIONS Concentration of 1 mg/ml of MSNc-PFP had a good Yuzhong district, Chongqing, China effect on the surface models of synergistic HIFU ablation. Correspondence: Yang S. Ying Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P25 P27 OBJECTIVES To explore the effect of the target blood vessels angled Effects of subatmospheric pressure and dissolved oxygen with acoustic axis on the formation of the around coagulation necro- concentration on the generation of lesions in ex vivo bovine livers sis under the model of the surface ablation of PHIFU with the same by HIFU conditions of irradiation dose, To adjust the HIFU treatment model Min He, Zhiqiang Zhong, Xiaobo Gong, Faqi Li, Deping Zeng, has certain guiding significance. Zhibiao Wang METHODS Embedded the rabbit thoracic aorta (diameter was 4.25 College of Biomedical Engineering, Chongqing Medical University, ±0.79 mm) into egg white body model and divided into 3 groups by Chongqing, China the blood vessels Angled 0 °,45°,90 °with acoustic axis.The egg white Correspondence: Min He body model were exposed to pHIFU with 200-300 w acoustic Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P27 power,100Hz pulse recurrence frequency(PRF),50% dutycycles.The surface ablation size is 30 * 30 * 30 mm3,line scanning, Linear speed: OBJECTIVES This study was aim to investigate the effects of subat- 3 mm/s;Line length: 30 mm; line scanning time interval: 1.5 min;Layer mospheric pressure and dissolved oxygen concentration on the spacing 2 mmand 16 layers in total.In the process of irradiation ultra- morphology and size of lesionsgenerated by HIFU in ex vivo bovine sonographic observations and temperature measurement; After the liver. irradiation, cut body model to observe theirradiation damage. METHODS All HIFU experiments were conducted in a stainless cham- RESULTS In the process of Irradiation, near the acoustic source side of ber fulfilled with degassed water. A 1MHz HIFU transducer was used the target blood vessels Angled with acoustic axis, the temperature to generate the US exposureof 11700W/cm2 which was acutely cavi- curve shown as: increase fast, long duration and falling fast and far tation under atmospheric pressure. The dissolved oxygen concentra- from the acoustic source side one is relatively slow, short duration, de- tion (DOC) of degassed water were divided into three groups: 1.0 creased slower. And the ultrasonographic performance that far from mg/L, 1.5 mg/L and 2.0 mg/L respectively. The ex vivo bovine livers the acoustic source side is low echo immediately, gradually enhanced, were exposing 2 seconds per time under two ambient pressure of at- after 3 min the gray level change is not obvious and near the acoustic mospheric pressure and subatmospheric pressure. B mode US moni- source side is immediatestrong echo, gradually weakened and after 3 toring the strong echo signal before and after HIFU exposing. A min the gray does not change significantly.After irradiation, cut down passive cavitation test system (PCD) test the acousticcavitation signal the egg white body models,observed that far from theacoustic source in the process of exposing. side of the blood vessel Angled 45 °, 90° with acoustic axis were no RESULTS 1. The broadband noise of atmospheric pressure and subat- white coagulation necrosis formed. And the 0°one coagulation necrosis mospheric pressure under the same dissolved oxygen concentration around the blood vessels was narrowed. shows that cavitation was weaker under subatmospheric pressure CONCLUSIONS Adopted the surface ablation model, the existence of than that under atmospheric pressure.2. The variation of difference the target blood vessels angled with acoustic axis can affects its sur- of gray level value between before and after HIFU exposing on the rounding coagulation necrosisformation. The target blood vessels An- B-scan image is increasing with the increase of dissolved oxygen gled 0 °with acoustic axis can narrowed the coagulation necrosis concentration. Under atmospheric pressure, the difference of gray arround the vessels. And far from the acoustic source side of the- level value is larger than that under subatmopheric pressure.3. Under blood vessels Angled 45 °,90 °could not formed the coagulation ne- atmospheric pressure, the lesions are from tadpole shape to elliptical crosis. And in the process of Irradiation, the change of the gray-scale as the dissolved oxygen concentration decreased. But under subat- ultrasonography could judge thedamage formation around the blood mopheric pressure, thelesions are all elliptical. The size of lesions is vessels to a certain extent. increasing with the increase of dissolved oxygen concentration. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 71 of 122 Under the same dissolved oxygen concentration, the size oflesions is RESULTS Effective follow-up were performed in 297 cases. The fol- larger under atmospheric pressure than that of subatmospheric low-up time was 1—50 months. There were 177 cases in diffuse pressure. group, while 120 cases in focalgroup. The incidence of ablation CONCLUSIONS This study investigated the effects of subatmospheric among 297 patients was 99.33% (295/297). The NPVR of diffuse and pressure and dissolved oxygen concentration on the generation of focal group were ([26.00±13.36] %) and ([44.32±19.93] %), respect- lesions in ex vivo bovine liversby HIFU. The followings were clarified ively. The dysmenorrhea and menorrhagia score of post-procedure in the experiment.1. The reduce of dissolved oxygen concentration were significantly lower than those of pre-procedure in two groups could decrease the volume of lesions generated by HIFU.2. Subatmo- (both P<0.05). The total remission rate of dysmenorrhea in 3, 6, 12, spheric pressure could restrain the cavitation, thus reshape the le- 24 and 36 months after ablation were 92.96% (264/284), 86.18% sions to elliptical and smaller the size of lesion in focus. (237/275), 73.51% (197/268),60.71%(136/224), 46.83% (59/126), re- spectively. The remission rate of dysmenorrhea of focal group was higher than that of diffuse group in each follow-up period, while sig- P28 nificant differences were observed in 6, 24 and 36 months (P<0.05). The effect of phased-hifu with discontinuous operating mode on The total remission rate of menorrhagia in 3, 6, 12, 24 and 36 months coagulative necrosis region after ablationwere 87.38% (187/214), 83.09% (172/207), 68.63% (140/ Xiongfei Qu, Guofeng Shen, Nan Wu, Yazhu Chen 204), 63.64% (105/165), 45.92% (45/98), respectively. The remission School of Biomedical Engineering, Shanghai Jiao Tong University, rate of menorrhagia of focal group washigher than that of diffuse Shanghai, China group in each follow-up period, while no significant difference was Correspondence: Xiongfei Qu observed between two groups (all P>0.05). Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P28 CONCLUSIONS HIFU is significant effective for adenomyosis, the short-term efficacy of focal and diffuse type are familiar, while the OBJECTIVES In the previous experiments, we noticed that Phased- long-term efficacy of focal typeis better than that of diffuse type. HIFU with discontinuous operating mode (eg. 2s heating following with a 1s cooling, and repeating) produced shorter and thicker co- agulative necrosis region in ex vivo porcine muscle, compared with slender spindle like region in continuous operating mode. The aim of P30 this study was to demonstrate the mechanism and influence of this Ultrasound-guided versus mr-guided high intensity focused method on tissue ablation. ultrasound for ablation of uterine fibroids by ultrasonic contrast METHODS In this study, we investigated the influence of discontinu- agent: treatment efficacy, safety and efficiency ous operating mode on tissue ablation. Three different treatment Yi Wang, Yonghua Xu procedures in 4 repeat cycle (each for 2.35s) were simulated using a Chongqing Medical University, The Institute of Ultrasound Engineering DFDT method: (1) Continuous heating with constant 200 watts of in Medicine, Chongqing, China acoustic power; (2)50% duty ratio (1.175s heating in a cycle) heating Correspondence: Yi Wang with 200 watts of acoustic power; (3)50% duty ratio heating with 400 Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P30 watts of acoustic power. Then the ex vivo porcine muscle and tissue- mimicking phantom (NIPAMbased hydrogel phantom with cloud OBJECTIVES The time efficiency and sonication energy efficiency point temperatures at 52°C) heating experiments were performed in of treatment and safety were compared between MR-guided procedures (1) and (3), to investigate the shape of coagulative necro- High-Intensity Focused Ultrasound (MRgHIFU) and Ultrasound- sis region and temperature above 52°C, respectively. guided High Intensity Focused Ultrasound (USgHIFU) for complete RESULTS The simulation showed that the short rod-like 240EM re- ablation of T2 hypointense fibroids. gion of the discontinuous operating mode was shorter, thicker and METHODS The treatment data and sonication parameters from 13 larger, compared with slender spindle like 240EM region of continu- uterine fibroids in 10 patients treated with MRgHIFU and 28 uter- ous operating mode. However, the gradient of thermal dose in ine fibroids in 22 patients treated with USgHIFU were analyzed. 240EM region boundary of discontinuous operating mode was much All of the pre-treatment fibroids were hypointense signal in T2 smaller, which may cause the instability of ablation border. These weighted imaging and completely ablated by using MRgHIFU two simulation results were observed in the ex vivo porcine muscle orUSgHIFU at one session treatment. The volume of the treated and tissue-mimicking phantom experiments. fibroid and the non-perfused volume (NPV) was calculated in CONCLUSIONS The discontinuous operating mode can produce a lar- contrast enhance MRI (CE-MRI), andcomplete ablation of fibroids ger short rod-like coagulative necrosis region, therefore may reduce is defined as non-perfusion region covering all volume of the the number of treatment shotsand improve the efficiency of treatment. treated fibroid immediately following the procedure. The treat- However, it reduces the gradient of thermal dose in ablation boundary, ment and sonication time, the EEF and NPV ratios were com- therefore may reduce the stability of coagulative necrosis region. pared between MRgHIFU and USgHIFU, while the adverse events and complications were also assessed. RESULTS The percentage rates of the completely ablated fibroids in P29 the MRgHIFU and USgHIFU were 29.5% and 41.2%, respectively. The Short-term and long-term efficacy of ultrasound ablation for treatment time was174.5±42.2 minutes and 114.4±39.2 minutes, the diffuse and focal adenomyosis sonication time was 24.7±9.0 minutes and 19.4±6.8 minutes, the son- Yujie Feng, Jinyun Chen ication power was 310.2±62.5W and 391.6±16.6W, the sonication en- College of Biomedical Engineering, Chongqing Medical University, ergy was 483.0±248.2 KJ and 463.2±156.4KJ, EEF was 5.1±3.0 KJ/cm3 Chongqing, China and 6.8±5.2 KJ/cm3, and the mean treatment speed was 42.2±25.6 Correspondence: Yujie Feng cm3/h and70.9±41.9 cm3/h in the MRgHIFU and USgHIFU treatment Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P29 for complete ablation of fibroids, respectively. There was a negative linear correlation between the EEF andthe NPV of fibroids, and a OBJECTIVES To compare the short-term and long-term efficacy of HIFU positive linear correlation between the treatment speed and the NPV in treatment of diffuse and focal adenomyosis. of fibroids in the both groups (P<0.05). There was a positive linear METHODS A total of 308 patients with adenomyosis who accepted correlation between the sonication intensity and the NPV of fibroids HIFU ablation were collected. According to preprocedural MRI, the in the USgHIFU group (P<0.05) and no correlation in the MRgHIFU patients were divided into diffuse and focal group. The non-perfused group (P>0.05). There was no severe adverse event and major com- volume ratio (NPVR) and the incidence of ablation were calculated. plication in both groups after treatment. Preprocedural and postprocedural situation of dysmenorrhea and CONCLUSIONS MRgHIFU and USgHIFU both are feasible, safe, and ef- menorrhagia were evaluated. fective with the equivalent energy efficiency for complete ablation of Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 72 of 122 T2 hypointense fibroids;USgHIFU was superior to MRgHIFU in the [3] Younan et al. "Influence of the pressure field distribution in transcranial time efficiency. ultrasonic neurostimulation." Medical physics (2013) [4] Ye et al. "Frequency Dependence of Ultrasound Neurostimulation in the Mouse Brain." Ultrasound in medicine & biology (2016) P31 [5] Li et al. "Improved Anatomical Specificity of Non-invasive Neuro- Estimation of thermal rise during ultrasonic neurostimulation in stimulation by High Frequency (5 MHz) Ultrasound." Scientific reports rodents: retrospective analysis of five recent studies (2016). Charlotte Constans, Mickael Tanter, Jean-Francois Aubry [6] Yoo et al. "Transcranial focused ultrasound to the thalamus alters Institut Langevin, Paris, France anesthesia time in rats." Neuroreport (2011) Correspondence: Charlotte Constans [7] Kamimura et al. "Focused ultrasound neuromodulation of cortical and Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P31 subcortical brain structures using 1.9 MHz." Medical Physics (2016) [8] B. Cox et al, k-space propagation models for acoustically heterogeneous OBJECTIVES The first ultrasonic neuromodulation studies were con- media: Application to biomedical photoacoustics, J. Acoust. Soc. Am., ducted with rather high intensity [1] and thermal effects were as- 2007. sumed to be the main cause [2]. More recently, multiple groups have [9] Burgoon et al. "Temperature-sensitive properties of rat suprachiasmatic reported successful low intensity focused ultrasound (LIFU) neurosti- nucleus neurons." American Journal of Physiology-Regulatory, Integrative mulation on rodents: movement elicitations [3] [4] [5] [7] or reduction and ComparativePhysiology (2001) of anesthesia time [6]. Given the low intensities used in most of them, mechanical effects are more prone to induce neuromodulation Table 1 (abstract P31). See text for description [4]. The mechanism of neuromodulation is still not fully understood, and a more thorough study of the thermal and mechanical effects is necessary to optimize the parameters for clinical applications. We simulated the thermal rise in 5 rodent studies in order to evaluate its potential impact. METHODS The acoustic propagation of focused ultrasound was simu- lated in an entire rat head in order to investigate the pressure ampli- tude and spatial distribution. The simulations were performed with k- Wave [6], a k-space pseudospectral method-based solver. 3D maps of the skull, brain and tissues were extracted from a rat microcomputed tomography scan. Brain and tissues were assumed to have the same sound-speed and density as water, and the transducer was modeled according to each study’s materials. Absorption was taken into ac- count in the skull (2.7dB/cm/MHz) and in the brain (0.21dB/cm/MHz) with a 1.18 power law of frequency. Ultrasound propagate in a cone filled with water before entering the rat head, the geometrical focal point being located about 7mm deep from the surface, inside the brain. For each study, we calculated the pressure at focal spot in water based on each study materials and methods. The simulations were first performed in water and compared to these extracted am- plitudes. The peak negative pressure in the rat head was extracted from the simulation and thus takes into account reflections and ab- sorption effects. The thermal code is based on the bio-heat equation without perfusion and metabolic processes. The thermal dose unit is Fig. 1 (abstract P31). Maximum temperature in brain, with zoom CEM (cumulative equivalent minutes at 43°C). on individual bursts (left) and at focal spot (right) in study [3] RESULTS Parameters and results in brain and at the focal spot for all the studies are listed in Table 1. Temperature rise estimated for [3] and [7] are plotted on Figs. 1and 2 respectively for the most heated point in the brain (left) and at the focal spot (right). In study [3], the thermal does not exceed 3.3E-4 CEM in the skull, brain and skin. In study [7], TD reaches 2800 CEM in the skull, 60 CEM in the brain (close to thebone) and 50 CEM in the skin. The thermal dose in the skin is significant but not high enough to induce skin burns. CONCLUSIONS Our retrospective analysis shows thermal rise ranging from 0.002°C to 9.3°C in the brain. For studies [3-6], corresponding to a frequency range of320kHz to 5MHz and a total sonication time ran- ging from 80ms to 20min, the maximum temperature elevation in the rodent brain is lower than 0.1°C. Sensitivity to temperature changes was found with a coefficient of 1.1 impulses/s/°C in some neurons [9]. Thus, in the case of study [7], the thermal rise cannot be Fig. 2 (abstract P31). Maximum temperature in brain, with zoom neglected as apossible cause of neuromodulation. on individual bursts (left) and at focal spot (right) in study [7] Acknowledgements This work was supported by the Bettencourt Schueller Foundation and the "Agence Nationale de la Recherche" under the program “Future Investments” P32 with the reference ANR-10-EQPX-15. Numerical simulation of the effect of phase transformation on standing waves and transcranial focusing in HIFU 1 1 2 1 References Miaomiao Zeng , Shihui Chang , Rui Cao , Xiqi Jian [1] Fry et al. "Production of reversible changes in the central nervous system Biomedical engineering, Tianjin Medical University, Tianjin, China; by ultrasound." Science (1958) Tianjin University of Science and Technology, Tianjin, China [2] Lele et al. "Effects of focused ultrasonic radiation on peripheral nerve, Correspondence: Miaomiao Zeng with observations on local heating." Experimental Neurology (1963) Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P32 Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 73 of 122 OBJECTIVES Reflections induced by heterogeneous structures and P33 large differences in acoustic impedance between the skull and peri- Analysis and investigation of the major parameter effecting enchyma result in undesired standing waves that cause energy loss transcranial ultrasound phase aberrations: a preliminary study in the treatment area in high-intensity focused ultrasound (HIFU) Nan Wu, Guofeng Shen, Xiongfei Qu, Yazhu Chen transcranial treatment and deposition of excess energy in healthy tis- School of Biomedical Engineering, Shanghai Jiao Tong University, sue. The goal of this work is to address these issues. Shanghai, China METHODS The simulations performed in this paper were based Correspondence: Nan Wu on the three dimensional finite difference time domain (FDTD) Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P33 simulation of acoustic propagation and thermal behavior through the bone windows. The material properties of the skull were de- OBJECTIVES Focused Ultrasound (FUS) is a noninvasive medical rived from 3D reconstructions of high-resolution computed tom- technology used for transcranial therapeutic applications. How- ography (CT) scans of selected patients. Phase transformation ever, due to the complex acoustic properties of the skull, it is methods were used to reduce the standing wave by randomly practically hard to obtain a sharp transcranial focus without changing the phase in time segments. phase correction. The aim of this research was to analyze and in- RESULTS The intensity of the standing wave decreased (Fig. 1). vestigate the major parameter of the skull which has the greatest Meanwhile, the sound pressure rose and the rate of temperature rise effect on phase aberrations when the ultrasound propagates at focal region increased when using phase transformation. Different through the skull. The correction of this parameter may simplify bone windows exhibit different optimum excitation frequencies, in the method of ultrasonic phase correction. the range of 0.6-0.8MHz. The minimum standing wave intensities ap- METHODS The numerical model was based on the k-wave simula- peared when the ratio of the phase transformation frequency to the tion environment, which was extensively tested and actually be- excitation frequency was 0.3 for all bone windows (Fig. 2). ing used to simulate the ultrasound field before. The whole CONCLUSIONS The phase transformation method has been proved simulation environment was designed in water, and a digitized to be effective in suppressing standing waves through variety bone human skull profile, which was built from CT (computed tomog- windows. The advantage of this method is that it can enhance the raphy) images, was placed below the transducer. To investigate energy of focal region and reduce the intensity of standing waves. how each parameter of the skull effects phase aberrations, dens- The optimum excitation frequencies selected for different bone win- ity, attenuation, velocity and geometry of the skull were taken dows were obtained and it was confirmed that the excitation fre- into account individually. The wave propagation simulations were quency and phase transformation frequency were relevant. performed with one of the skull parameters, hypothesizing the others set to be the parameters of water throughout the simula- tion process. The benchmark configuration for this research was the spherically curved transducer driven at 700 kHz; the trans- ducer had a curvature radius of 120 mm and a diameter of 90 mm. The focus acquired from different parameters was compared with each other after the simulations. RESULTS A transducer placed in a homogeneous media (water), without and with the skull were simulated to be comparison groups. The focus of the transducer was shifted and defocused when the skull was placed in the sound field. However, a sharp focus still could be achieved when the density or attenuation of the skull was taken into account. An aberrant focus was gener- ated when the velocity was set to 2850m/s (velocity of the skull). It was interesting to note that, the more velocity vectors penetrate the skull perpendicularly, the better a focus could be Fig. 1 (abstract P32). The acoustic pressure distribution (temporal obtained. bone window, 0.6 MHz, I =4.0 Wcm , t =16 s). (a) without and (b) CONCLUSIONS This research presented a simulation to analyze with standing wave suppression and investigate the major skull parameter effecting transcranial ultrasound wave phase aberrations. The velocity of skull could be the major parameter on phase aberrations, compared with the density and attenuation. Besides, the more velocity vectors per- pendicular to the skull, the less phase aberrations and higher soundpressure couldbeobtained. Amethodonmakingmore velocity vectors penetrating the skull perpendicularly that can simplify the correction of ultrasound wave phase will be done in the further research. P34 Protective effect of ultrasound on brain injury in mice Feng-Yi Yang, Wei-Shen Su Department of Biomedical Imaging and Radiological Sciences, National Yang-Ming University, Taipei, Taiwan Correspondence: Feng-Yi Yang Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P34 OBJECTIVES The purpose of this study was to investigate the effects of low-intensity pulsed ultrasound (LIPUS) in mice with traumatic brain injury (TBI). METHODS Controlled cortical impact (CCI) injury was used as a TBI animal model. Mice subjected to CCI injury were treated with LIPUS Fig. 2 (abstract P32). Plot of Rα against fr./f (temporal scales area at daily for a period of 28days. Behavioral assessments (mNSS and 0.6MHz, occipital area and parietal area at 0.7MHz) rotarod) were performed at day 28 after TBI. Histological examination Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 74 of 122 was performed via cresyl violet staining. Edema regions were moni- normalized therapeutic gain plots for the lateral and elevation direc- tored by magnetic resonance imaging. tions. The hydrophone measured pre-focal planes showed strong RESULTS Our data showed that functional impairments were signifi- agreement with back propagated simulations. As the frequency is in- cantly improved by LIPUS stimulation at day 28 after TBI. LIPUS sig- creased, the distortion and attenuation increases. Figure 2 shows the nificantly preserved brain tissue compared with the non-treated TBI back propagation results for a single frequency of 3.2 MHz. When group. Furthermore, LIPUS significantly reduced T2-weighted lesion analyzed as a whole frequency set, they show sections of the skull hav- volume in injured mice compared with the non-treatedTBI group. ing different frequency propagation characteristics. Complex skull-bone CONCLUSIONS In summary, LIPUS stimulation improved long-term structures and suture lines increase focal distortion and decrease total behavioral outcomes and attenuated brain tissue damage in mice powertransmission. Contiguous regions display greater transmission subjected to TBI. Therefore, transcranial LIPUS stimulation may pro- with decreased distortion and foci shifting. Figure 3 shows the bregma vide a potential treatment modality for TBI. suture line (Plane 1) causes stronger distortions than transmission re- gions between the bregma and lambda suture lines. The grid of points behind the skull show the distortion is not constant throughout theb- rain. These distortions are dependent upon both frequency and the re- P35 gion of the skull the beam is traveling through to get to the respective Broadband characterization of focused ultrasound transskull imaging pixel. These results point towards creating a pre-treatment cal- transmission culation of the effectiveness of a successful therapeutic delivery. Parker D. O'Brien, Dalong Liu, Emad S. Ebbini CONCLUSIONS The results demonstrate the feasibility of DMUA im- Electrical and Computer Engineering, University of Minnesota–Twin aging (both SA and STF) to characterize the transmission loss Cities, Richfield, Minnesota, USA through the skull as can be seen from the general agreement with Correspondence: Parker D. O'Brien the hydrophone estimates (Fig. 3). The results also demonstrate that Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P35 a relatively wide transmission bandwidth in both rodent and human skulls. Based on these results, it would be possible to transmit broad- OBJECTIVES The future of transcranial Focused Ultrasound (tFUS) for band tFUS beams that can be designed to minimize the focal spot subtherapeutic (neuromodulation) or ablative treatments in the hu- and improve the specificity of both ablative and subtherapeutic (neu- man brain relies on spatially specific therapeutic delivery with quanti- romodulation) tFUS beams. fiable power control to enable reversible and irreversible treatments. Single frequency transmission signals may not provide adequately defined foci nor minimal foci shift after propagating through the varying complex structures of the skull to target fine neuro-structures within the brain. Broadband transmission is highly likely to provide the most effective method of delivering specialized therapeutic treat- ments obtainted through its ability to recover distortion and loss from a single frequency with a wide range of available frequencies. This paper presents the broadband transmission characterization of FUS through rat skulls and human skulls ex vivo through various re- gions of the skulls to understand how different frequencies can be used to refocus distorted or recover lost transmission through the use of multiple frequencies. METHODS Two dual-mode ultrasound arrays (DMUA) are used to transmit FUS through a series of human and rodent skulls. Both DMUAs are concave spherically focused, linear arrays with radii of curvature specific for the skull model used (r=100mm for human, r=40mm for rat) (Imasonic, France). Planar acoustic pressure mea- surements using needle hydrophones (Onda Corp, Sunnyvale, CA) were performed in degassed water with and without skull samples present (Fig. 1). Field scans were performed in prefocal and focal planes on a finely sampled grid with sub-wavelength spacing. Figure 1 shows the setup using a human skull in the water bath with the 600-micron hydrophone and DMUA (100-mm ROC) operating in the 0.7 - 2.5 MHz frequency range. Total power was estimated by inte- grating the square intensity (from hydrophone measurements). In Fig. 1 (abstract P35). See text for description addition, the focal plane measurements were back propagated to the interior of the skull to predict the shape of the acousticwavefront transmitted through the skull. Pre-focal field scans were performed to validate the backpropagation computational models between the focal plane and prefocal plane scans. Imaging distortion was charac- terized through a grid of single-point hydrophone measurements of synthetic aperture (SA) and single-transmit-focus(STF) imaging for the rat skulls with the 40-mm DMUA (operating bandwidth of 2.2 - 4.6 MHz). For all experiments, a range of frequencies relative to the human and ratmodel were chosen to accomplish a thorough broad- band characterization for all relevant frequencies (Human: 1.00,1.35,1.70,2.10 MHz; Rat: 1.80, 2.50,3.20,3.70,4.20MHz). RESULTS The hydrophone scans and backpropagation provide a thor- ough depiction of the transmission of focused ultrasound from the in- terior of the skull to the focalplane. The focal plane shows the Fig. 2 (abstract P35). See text for description distortion and loss of power caused by the skull. Figure 2 shows Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 75 of 122 the set point. All animals were survived for 3 - 5 days after tFUS applica- tion and observed for any abnormal behavior or adverse reaction. Histological evaluation was performed to determine whether the deliv- ered tFUS dose produced a BBB opening. For some animals, we ex- tracted the skull and performed in vitro transskull field mapping experiments to characterize the actual distortion to the tFUS beam in different planes with respect to bregma. RESULTS All animals that underwent the subtherapeutic tFUS applica- tions described above (over 30 Sprague Dawley rats 275 - 475 gm, male and female) have survived the procedure and no adverse events were recorded. Figure 2 shows an example of the rendering of the 3D im- aging data using the DMUA render engine. Onecan see the C-mode view with the lambda and bregma suture lines clearly visible with a thick arrow pointing to the bregma. The lines show the selected planes for tFUSapplication in a typical experiment. Figure 3 demonstrate the Fig. 3 (abstract P35). See text for description feasibility to visualize the tFUS beam access by rendering the DMUA el- ements and the skull (to scale) by using the results from the 3D scan. This allows for pre-treatment planning and post-treatment evaluation by bringing computational modeling and DMUA imaging data together in one computational model. Figure 4 shows an example result from a P36 spatiotemporal control of tFUS-induced temperature rise of 4°C for 10 Real-time spatiotemporal control of transcranial focused sec using the setup in Fig. 1. The pseudocolor overlay shows the spatial ultrasound in vivo distribution of the heated region and the spatiotemporal map (axial- 2,1 1 Dalong Liu , Emad S. Ebbini temporal) shows the localization in the axial direction. Electrical and Computer Engineering, University of Minnesota, CONCLUSIONS The results shown demonstrate the feasibility of real- Minneapolis, Minnesota, USA; Siemens, Seattle, Washington, USA time precise spatiotemporal control of tFUS dose application in a Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P36 temperature control application (e.g. drug-delivery BBB opening), but the DMUA could be easily used to control nonthermal application of Correspondence: Dalong Liu tFUS (e.g. neuromodulation). The results also demonstrate the unique Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P36 advantages of the DMUA approach, where DMUA imaging data pro- vides 3D volumetric rendering of the skull for target localization, pre- OBJECTIVES Transcranial focused ultrasound (tFUS) has been receiv- treatment planning and post-treatment evaluation. We envision a ing increasing attention by numerous research groups worldwide. It prescription-tFUS application using DMUA technology. is being investigated as a potential noninvasive treatment modality for tumor ablation, drug delivery through blood-brain barrier open- ing, Parkinson disease, etc. This renewed interest in tFUS canbe cred- ited to advances in diagnostic imaging and image-guidance modalities, especially MRI. The distortion of tFUS beam through the skull continues to be a major challenge to the ultimate goal of reli- ably localizing and controlling the therapeutic tFUS dose to meet the demands of precision therapy of neurological disorders. The goal of this study is to establish the feasibility of precise real-time spatiotem- poral control of tFUS energy application in a rat model in vivo. METHODS The 3.5-MHz dual-mode ultrasound array (DMUA) prototype was used to deliver tFUS to anesthetized animals under IACUC-ap- proved protocol. In each experiment, the animal was positioned prone on a stereotaxic unit with the head shaved and hair removed using de- pilatory cream to allow for good coupling with theDMUA through its water bolus (Fig. 1). A 3D scan of the skull was performed by mechanic- ally translating the DMUA prototype using a 3-axis motor (caudal-to- Fig. 1 (abstract P36). The dualmode ultrasound array (DMUA) used rostral). This scan was performed to identify the lambda and bregma in imageguided spatiotemporal control of tFUS in rat model in vivo suture lines, which were used a a reference for the treatment planes. The DMUA system provided a 3Drender interface to allow the visualization of these markers upon the completion of the 3D scan and before starting the tFUS application in selected planes (based on the- target circuitry for a given application.) All treatment planes were marked with respect to the bregma (e.g. bregma-2mm). Once the DMUA imaging/treatment slice was aligned with the desired treatment plane, imaging was performed before, during and after the application of the tFUS dose. Our system allowed for a full range of thermal and nonthermal control of tFUS by controlling the duty cycle. Spatial con- trol of the tFUS beam was provided by precision refocusing using a multichannel arbitrary waveform generation driver. In addition, ampli- tude modulation of tFUS on a frame-by frame basis (up to 500 fps) was achieved using a closed-loop control system basedon DMUA feedback. For the purposes of this paper, we describe the spatiotemporal control of subtherapeutic tFUS beams for thermal therapy. A typical tFUS shot wasbetween 4 - 15 seconds with initial exposure of approximately 400 W/cm2 in situ, designed to reach the temperature set point in ~0.5 sec. Fig. 2 (abstract P36). The 3D render interface used for visualization Ultrasound thermography basedon DMUA beamformed echo data of the bregma suture line (thick arrows) as reference for the from the target region was used for feedback. A PID controller was treatment planes (lines) employed to adjust the tFUS intensity to maintain the temperature at Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 76 of 122 Currently, there is no cure for PD patients and only a few options are available to alleviate symptoms. With advances in gene therapy re- search, adeno-associated virus (AAV) has the potential to serve as carrier to introduce therapeutic genes to the human body. In con- trast to invasive direct brain infusion, focused ultrasound (FUS) in combination with microbubbles has been shown to non-invasively and transiently open the blood-brain barrier (BBB). Here, our goal is to evaluate the potential neuro-protective and neuro-restorative ef- fects of non-invasive AAV-GDNF delivery in a MPTP mouse model. METHODS The PD mouse model was generated via intraperitoneal injec- tions of MPTP toxin at 30 μg/kg over five consecutive days. Animals were then divided into four groups (n = 7-10 per group): control, FUS only, AAV injection only, and FUS+/AAV+. For the FUS only and FUS+/AAV+ groups, both striatum and substantia nigra were sonicated unilaterally using a single element FUS transducer. For the AAV+/FUS+ group, a 100 μl mixture of AAV-GDNF vectors and polydispersed microbubbles were administered intravenously before sonication. Mice were allowed to sur- vive up to three months’ post sonication, which was followed by transcar- dial perfusion and tissue analysis. RESULTS Systemically administrated AAV vectors were capable of crossing the opened BBB and viral transduction was observed to be concentrated at the FUS targeted brain regions (i.e. striatum and sub- stantia nigra). The number of dopaminergic neurons at the point of sacrifice for each mouse was quantified by staining for tyrosine hy- droxylase (TH) in the substantia nigra regions. As shown in Fig. 1, mice that received a combination of AAV-GDNF and FUS exhibited Fig. 3 (abstract P36). A 3D isometric view of the DMUA elements significantly higher protection to the subsequent MPTP insult. In and the imaging/treatment slice with respect to the skull (as addition, neurorestoration was observed in AAV+FUS treated mice rendered from DMUA 3D imaging data) that was previously dosed with MPTP toxin. The dopaminergic neuron projections on the FUS+/AAV+ hemisphere also had higher density than the contralateral side. Behavioral study was performed 12 weeks after the initial unilateral treatment, where amphetamine- elicited unilateral rotation was observed in mice from the combined (AAV+FUS) treatment group (p < 0.05). CONCLUSIONS The findings of this study indicate the potential of gene delivery vectors for protecting and restoring the functions of dopaminergic neurons in PD and FUS as a non-invasive methodology for transcranial AAV delivery. Fig. 1 (abstract P37). a) Immunofluorescence TH staining revealed much more dopaminergic projections on the AAV+/FUS+ side of Fig. 4 (abstract P36). Typical result of realtime spatiotemporal the brain. b) Significantly higher number of TH+ neurons were found control of subtherapeutic tFUS thermal application in vivo rat model on the AAV+/FUS+ side of the brain compared to the contralateral side. c) Significantly higher dendrite density was observed on the AAV +/FUS+ side of the brain. d) Amphetamine-elicited behavioral studies revealed more frequent clockwise (toward the remaining lesion side) P37 rotation, signifying more prominent dopaminergic function on the Focused ultrasound-facilitated AAV-GDNF delivery for neuro-protection hemisphere receiving AAV+/FUS+ treatment and neuro-restoration in Parkinson ’s disease mice 1 1 1 Shutao Wang , Oluyemi Olumolade , Tara Kugelman , 2 1 2 Vernice Jackson-Lewis , Maria Eleni Karakatsani , Serge Przedborski , 1,3 P38 and Elisa E. Konofagou Transcranial focal passive detection and imaging: implementation Department of Biomedical Engineering, Columbia University, New York, on a multiple channel transmit/receive ultrasound phased array NY, USA; Department of Neurology, Columbia University, New York, NY, system USA; Department of Radiology, Columbia University, New York, NY, USA Chih-Hung Tsai, Hsin-Yu Chang, Chung-Han Wang, Hao-Li Liu Correspondence: Shutao Wang Department of Electrical Engineering, Chang Gung University, Taoyuan Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P37 City, Taiwan Correspondence: Chih-Hung Tsai OBJECTIVES Parkinson’s disease (PD) is the second most common Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P38 neurodegenerative disorder affecting millions of patients worldwide. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 77 of 122 OBJECTIVES Burst-mode focused ultrasound (FUS) exposure combined OBJECTIVES Low intensity focused ultrasound (LIFU) has the dem- with microbubbles (MBs) has been shown to induce temporal and local onstrated ability of non-invasively stimulating neural activity. This blood-brain barrier (BBB) opening. Contrast-enhanced imaging is now is of high value for therapeutic (stimulation, neuroprosthetics, served an indicator to postoperatively confirm the occurrence of BBB etc.) and diagnostic (preoperative mapping, etc.) purposes. A mul- opening. Developing a transmit/receive dualmode FUS apparatus has the tiscale simulation platform for image-based and personalized potential to observe focal position and fulfill implementation of real-time modeling of transcranial LIFU stimulation should be developed to monitoring of the occurrence of BBB opening. This study aims to disclose allow mechanistic studies, hypothesis formulation and testing, de- our recent development in using a self-designed multiple-channel trans- vice development, and, ultimately, personalized treatment plan- mit/receive system allow to perform passive cavitation analysis as well as ning, safety, and efficacy assessment. Experimental validation is to reconstruct focal beam distribution via passive imaging reconstruction. crucial to establish confidence in and explore the limitations of METHODS Homemade 256-channel ultrasound phased array driving sys- the modeling. tem was employed to drive a 256-channel FUS transducer to deliver focal METHODS The Sim4Life computational life sciences has been ex- transmit energy (fundamental frequency = 500 KHz, diameter = 120 mm, tended to: 1) Support full-wave acoustic simulation of transcranial curvature = 100 mm). The transmit pulse was designed to be 0.006 ms of sonication: For that purpose, new functionality to consider CT image- burst length, 2 Hz of pulse repeated frequency (PRF) and 3.56 MPa nega- based skull inhomogeneity information (density, speed-of-sound, and tive pressure output. Received circuits with the channel number ranging attenuation maps) and partly compensate for related focus aberra- from 16-64 were employed to perform RF signal receiving. During the in- tion and defocusing with multi-element transducer steering vitro experiments, either a strong needle reflector or microbubble (MB) optimization has been implemented. 2) Allow for neuronal dynamics tube was positioned with the flowed MBs concentrations been con- modeling: The NEURON library for compartmental neuronal dynamics trolled, the multiple channel RF signals were received in parallel with the modeling supporting detailed neuromorphology and channel dy- human skull were inserted. Passive cavitation detection was imple- namics has been integrated and parallelized simulations featuring mented, and passive imaging was reconstructed with the developed large numbers of neurons and neural networks can now be per- phase-corrected passive beamformed algorithm. formed. 3) Generate personalized, functionalized head models: Pa- RESULTS We demonstrate the feasibility in using this self-designed mul- tient image data can be used to generate anatomical geometries by tiple-channel system to serve as a platform to be operated at dual trans- segmentation and/or morphing of presegmented models. CT image mit/receive mode (Fig. 1). Multiple channels of RF data can be received in data informs on inhomogeneity, DTI image-data can be used for fiber parallel to reconstruct the passive imaging. The system now can support tracking to generate neuronal axon models, and Python tools facili- up to 64-parallel channel receiving for the following signal analysis and tate the anatomo-physiologically correct placement of cortical pyr- passive imaging formation. Point-spread function (PSF) imaging can be re- amidal neuron and deep brain stimulation relevant neurons (STN, constructed singly using 16-channel receiving, whereas higher channel GPi, IC). 4) Coupled acoustic and neuronal dynamics modeling: The provide superior SNR of imaging. The system also demonstrates the cap- Plaksin-Shoham-Kimmel (PSK) model of membrane-cavitation in- ability of the focal passive cavitation detection to real-time trace cavitation duced neurostimulation has been implemented and adapted for future activity specifically originating from the focal point. We also demonstrated use in combination with the compartmental cell models. Furthermore, that the implementation of a filtered phase-correction processing been coupled electromagnetic neuronal modeling is also supported. An ex- applied into the PSF reconstruction algorithm can successfully identify the perimental setup involving an acoustic transducer sonicating through a focal ultrasound deposition when penetrating through the skull. rat skull has been constructed. MicroCT image data has been acquired CONCLUSIONS We demonstrated the feasibility of the capability in and the 3D pressure distribution inside the skull has been measured using a self-built multiple-channel ultrasound transmit/receive system using computer controlled hydrophone scanning. to perform passive imagingand real-time focal PCD. The system and RESULTS The acoustic solver has been extensively validated previ- architecture has the potential to be developed to real-time monitor the ously against numerical and experimental data in homogeneous process of microbubble-facilitated FUS BBB opening process. setups and setups with homogeneous obstacles. The new experi- mental data allows for the first time successful validation of a setup involving inhomogeneous media using the previously pre- sented Gamma method for uncertainty assessment-based, object- ive comparison of 3D pressure distributions (Fig. 1).The neuron functionalized anatomical head models have been partly vali- dated by comparing modeling of transcranial electric/magnetic and deep brain stimulation with experimental data from literatur- e.The PSK-model could be simplified without significant impact on the results, thus enabling its integration into 3D, extended, morphological cell models. For that purpose, the previously bidir- ectional coupling of the electrophysiological and cavitation me- chanics parts has been broken up and further separation allows to pre-compute costly parts of the model, accelerating the mod- eling by more than one order of magnitude. Fig. 1 (abstract P38). Passive imaging reconstruction: upper row CONCLUSIONS A multi-scale framework for the computational inves- shows the tube was filled with air, and lower row shows tube was tigation of LIFU neuro-stimulation is being developed. It features filled with MBs image-based acoustic propagation modeling (including support for bone inhomogeneity) and focusing functionality, patient-specific functionalized anatomical model generation with realistic neuron P39 placement, integrated neuronal dynamics simulation, and a Multiscale Modeling of transcranial focused ultrasound coupled model of LIFU-induced neurostimulation that is currently neurostimulation and experimental validation: initial results still limited to 0D neuron models, but has been prepared for the fu- 1,2 2 1 3 Hazael Montanaro , Mehmet S. Özdas , Esra Neufeld , Théo Lemaire , ture modeling of 3D-extended, morphologically-detailed physio- 3 2 1, 2 Silvestro Micera , Mehmet F. Yanik , Niels Kuster logical neuron models. Important parts of the platform (acoustic Computational Life Sciences, IT'IS Foundation for Research on and neuronal simulations, functionalized models) could be success- Information Technologies in Society, Zurich, Zurich, Switzerland; Swiss fully validated against new and existing experimental data.The pre- Federal Institute ofTechnology (ETHZ), Zurich, Switzerland; Swiss Federal sented progress is an important step towards the goal of allowing Institute of Technology (EPFL), Lausanne, Switzerland mechanistic studies, hypothesis formulation and testing, device de- Correspondence: Hazael Montanaro velopment, and, ultimately, personalized treatment planning, safety, Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P39 and efficacy assessment. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 78 of 122 ofPSPIO and pDNA onto MBs were 134.5 g and 17.1 g, individually. Figure 1B shows that PSp-MBs with ultrasound could achieve gene transfection and the expression of gene could be further enhanced by MT process. We also confirmed that the PSp-MBs could perform successful BBB opening without bioeffects by the trigger ofultra- sound with an acoustic pressure of 0.3 MPa (Fig. 1C). CONCLUSIONS The MBs have fairly payloads of PSPIO-pDNA. We demonstrated that PSp-MBs with ultrasound can perform locally gene delivery and open BBB concurrently. In addition, the efficiency of gene delivery could be further enhanced by MT process. Future works include quantifying and tracking of distribution of gene deliv- ery and Parkinson’s disease rats via magnetic resonance imaging. Fig. 1 (abstract P39). Cortex subregion of a human model with anatomically positioned pyramidal neurons and visualized transmembrane voltage activity P40 SPIO-PEI-pDNA complex loaded microbubbles for ultrasound-based gene therapy in brain Chun-Yao Wu, Ching-Hsiang Fan, Rih-Yang Huang, Chien-Wen Chang, Chih-Kuang Yeh Department of Biomedical Engineering and Environmental Sciences, National Tsing Hua University, Hsinchu, Taiwan Correspondence: Chun-Yao Wu Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P40 OBJECTIVES Recently, gene therapy has attracted much attention espe- cially in neurodegenerative diseases. Currently, gene delivery within cen- tral nerves system mainly relies on invasive intracerebral injection or viral vectors to circumvent the obstacle of blood-brain barrier. Non-viral gene delivery via systematic transvascularroute is an attractive alterna- tive since it is non-invasive. However, a high-yield and targeted gene delivery platform is still lacking. In order to improve the efficiency of gene delivery, this study proposed an ultrasonic sensing vector for gene delivery into brain through polyethylenimine (PEI)-superparamagnetic Fig. 1 (abstract P40). (A) Microscope images of the PSpMBs with iron oxide (SPIO)-pDNAloaded microbubbles (PSp-MBs). Cooperating Prussian Blue and propidium iodide staining. The colocalization of with ultrasound exposure, PSp-MBs could transport the PSp nanoparti- the PSp nanoparticles in the Prussian Blue and propidium iodide cles into the desired brain region by acoustic MBs cavitation activity. The staining indicated a good conjunction of PSPIOpDNA and MBs. (B) rate of gene transfection would be enhanced by the modification of PEI Gene transfection efficency of PSpMBs with different conditions, onto PSp nanoparticles. In addition, by an externally applied magnetic suggesting that PSpMBs with ultrasound could achieve gene field, magnetic targeting (MT) can further increase the deposition of PSp transfection. (C) BBB disruption of PSpMBs cooperating with FUS by at the targeted location, enhancing the gene delivery. different parameters METHODS The PSPIO was consisted of PEI molecular and SPIO nano- particles (diameter: 10 nm) via ligand exchange. The PSPIO were then conjugated with pDNA and loaded onto the lipid surface of MBs by electrostatic force. PSPIO-pDNA (luciferase plasmid) modulated onto the MBs was confirmed by Prussian blue staining and propidium iod- P41 ide staining. The size, concentration and PSPIO payload were mea- Numerical study of bubble area evolution during acoustic droplet sured by multisizer and plate reader, respectively. C6 glioma cell and vaporization enhanced HIFU treatment 1 1 1 2 Sprague-Dawleyrats (N = 4) were used in this study. The gene trans- Ying Xin , Aili Zhang , Lisa X. Xu , Jeffrey B. Fowlkes fection efficiency and BBB opening region resulted from PSp-MBs School of Biomedical Engineering, Shanghai Jiao Tong University, with ultrasound (frequency = 1 MHz, energy = 0.1-0.5MPa, cycle = Shanghai, China; Department of Radiology, University of Michigan 5000, PRF = 1 Hz, sonication time = 60 s) were evaluated by bio- Health System, Ann Arbor, Michigan, USA luminescence imaging and Evans blue staining, individually. The MT Correspondence: Ying Xin process was performed by a 0.48 Tesla external magnet. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P41 RESULTS Figure 1A shows the fabricated PSp-MBs. The co- localization of the PSp nanoparticles in the bright field images and OBJECTIVES Acoustic Droplet Vaporization (ADV) has the potential to the fluorescent image indicated a good conjunction of PSPIO-pDNA shorten treatment time of high intensity focused ultrasound (HIFU) and MBs (as arrows). The mean size and concentration of PSp-MBs while minimizing thepossible effects of microbubbles along the were 1.5 m and (5-10) × 109 bubbles/mL, respectively. The payload propagation path. Distribution of the bubbles formed from the Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 79 of 122 droplets during the treatment is the major factor shaping the therapeutic region. However, there is no simulation report of this phase shift droplets assisted HIFU therapy, in which the bubbles form dynamically and only exists in places where certain acoustic conditions are met. In order to provide an approach for compre- hensive parametric analysis, which could save time and effort in future studies, simulation of the dynamic formation of the bub- bles inside the tissue during this treatment is carried out in this paper. The proposed model is verified using previously published experimental results. Numerical results predicting the effect of the presence of a preformed bubble wall are also obtained and discussed. METHODS The schematic of the model setup and the axisymmetric geometry of the simulation spatial domain are shown in Fig. 1. The Fig. 1 (abstract P41). (a) Schematic of the physical model and (b) origin of coordinate system coincides with the HIFU geometric focus. the geometry of the axisymmetric simulation spatial domain The phase shift droplets are assumed to be distributed uniformly in the phantom. The vaporized bubbles also distribute evenly, which is expected in phantoms and likely the case for larger bubbles trapped within the vasculature in tissue. The average diameters of droplets are usually smaller than 5 microns (Kripfgans et al., 2000; Zhangand Porter, 2010), and the formed bubbles have a diameter 5 times of the one of the droplets based on ideal gas law, which is much smaller than the ultrasound wavelength (about 1.9 mm to 0.5 mm when the frequency is in the therapeutic range of 750 kHz to 3 MHz), so the bubble area is treated as a homogeneous medium. Linear sound equation with attenuation in the frequency domain is used to describe the acoustic pressure p. The density ρ, sound speed c and attenuation coefficient α are functions of the gas void fraction fG, which can be calculated directly from the bubble size distribution (Church, 1995). ADV droplets would be vaporized under certain con- ditions, which are related to ultrasound frequency, diameter of the droplets, super heat degree and total ultrasound ‘ontime’. The follow- ing assumptions are used to decide how much bubbles will form in certain acoustic field:(i) Threshold for droplets in phantom gel ex- posed to 750 kHz ultrasound is found to be 7.6 MPa as was found in experiment.(ii) The probability of droplets vaporizing increases linearly with rarefactional pressure in a range, which is function of the threshold according to the experimental results (Lo, et al, 2007). (iii) The incident wave got reflected at the interface of the bubble Fig. 2 (abstract P41). (a) Droplets size distribution used in the area based on the difference of the acoustic impedance of the two simulation. (b) Sound speed and (c) sound attenuation coefficient medium. Meanwhile the bubble area presents different impedance as a function of void fraction in the phantom. (d)Sound speed as to different frequency. So the nonlinear part of the acoustic field is function of frequency (void fraction is 2.52×104 mL/mL) considered and simplified based on it. RESULTS The ultrasound used in the model is 750 kHz, with a focal pressure of 9.8 MPa and 14.7 MPa. Each pulse is 20 μs. The droplets concentration is 1.3×106/mLand the size distribution of the droplets is shown in Fig. 2(a). The calculated sound speed and sound atten- cuation is shown in Fig. 2(b-d) The sectional view of pressure field and bubble area after 20 cycles of ultrasound exposure when focal pressure is 9.8 MPa is shown in Fig. 3. The sectional view of pressure field and bubble area after 200 cycles of ultrasound exposure when focal pressure is 14.7 MPa is shown in Fig. 4 (a, b). All the pressure was representing in dB scale using 9.8 MPa as the reference pressure. These results are close to the experimental result (Lo. at al, 2006) that they can describe the unique feature of the bubble area when the in- cident pressure varies. A layer of bubble has been proved to provide protection of the distal area in HIFU therapy. This model can also predict the bubble area when a bubble layer is created before treat- ment in the therapy (Fig. 4 (c, d)). CONCLUSIONS This model can describe the bubble area evolution during the acoustic droplet vaporization enhanced HIFU therapy Fig. 3 (abstract P41). The sectional view of (a) pressure field and (b) while using a simple linear acoustic model. The model can also be bubble area after 20 cycles of 20 μs pulses of ultrasound exposure, used in the case when a bubble layer pre-exists to protect the distal of which focal pressure is 9.8 MPa and frequency is 750 kHz. The area from the HIFU ablation. The heating effect can be coupled to pressure was represented in dB scale using 9.8 MPa as the reference the model in the future and then it can be used as a planning tool in pressure. The white scale bar represents 2 mm ADV enhanced HIFU therapy and predict the treatment outcome. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 80 of 122 different FUS pulses ([peak-negative pressure (Pneg): 8, 12MPa, pulse length (PL): continuous wave], [Pneg: 25MPa, PL: 5 ms, PRF: 20 Hz]) wereemitted. Dye movement (i.e., clearance) at the hydrogel surface was observed with a video camera. METHODS Our initial challenge was to create an appropriate material for studying acoustic streaming. We hypothesised that existing hydrogels (e.g., polyacrylamide and gelatin) used as ultrasound phan- toms, mimic the acoustic properties of tissue, but not the tissue microenvironment. We created three phantoms – gelatin, polyacryl- amide (PAA), and a new macroporous polyacrylamide (MPPA) hydro- gel – and analysed their structure using scanning electron microscopy (SEM) and mercury intrusion porosimetry (MIP). Based on our findings, we selected the MPPA for our tissue-mimicking material. Focused ultrasound (FUS) pulses were emitted from a single-element transducer [centre frequency (fc): 5 MHz], which was driven by a function generator through a power amplifier. The focal point of a FUS transducer’s beam (axial FWHM: 3.2 mm, lateral FWHM: 0.45 mm) was placed at the distal surface of the MPPA. A model drug (Bromophenol blue) was injected in and around the focal region and different FUS pulses ([peak-negative pressure (Pneg): 8, 12MPa, pulse length (PL): continuous wave], [Pneg: 25MPa, PL: 5 ms, PRF: 20 Hz]) wereemitted. Dye movement (i.e., clearance) at the hydrogel surface was observed with a video camera. RESULTS SEM revealed that although gelatin and PAA were porous, they Fig. 4 (abstract P41). The sectional view of (a) pressure field and (b) had pockets of water separated by walls of material (Fig. 1a, b). This sug- bubble area after 200 cycles of 20 μs pulses of ultrasound exposure, gests that acoustic streaming with these materials are unlikely to occur of which focal pressure is 14.7 MPa andfrequency is 750 kHz. The unless enough stress is applied to break the walls. In contrast, MPPA had sectional view of (c) pressure field and (d) bubble area when a 1 interconnected pores throughout the gel – a feature that more accurately mm thick of bubble layer pre-existed. The pressure was represented resembled the interstitial space of soft tissue (Fig. 1c). The gelatin, PAA in dB scaleusing 9.8 MPa as the reference pressure. The white scale and MPPA had a porosity of 81, 76 and 88%, respectively; a permeability bar represents 2 mm of 2400, 1500 and 6500 mdarcy, respectively. We then selected MPPA for further study, because it most accurately mimicked the tissue microenvir- onment. Using focussed ultrasound in continuous wave (Pneg: 12MPa), P42 we directly observed acoustic streaming in the MPPA material in the form Noninvasive and localised acoustic micropumping – an in vitro of the dye progressively clearing from the focal region (Fig. 2). The clear- study of an ultrasound method that enhances drug distribution ance rate is represented by the change in optical density over time (Fig. through a physiologically-relevant material 1 1 2 3).Pulsedultrasoundwasapplied(Pneg:25MPa,PL:5ms,PRF:20 Hz) Ahmed Elghamrawy , Florentina de Comtes , Hasan Koruk , 3 1 having the same average intensity as 8MPa continuous wave and an in- Ali Mohammed , James J. Choi crease in clearance rate was observed. Moreover, pulsed ultrasound re- Department of Bioengineering, Imperial College London, London, UK; 2 3 ducedunwantedthermaleffects. Mechanical Engineering, MEF University, Istanbul, Turkey; Materials, CONCLUSIONS Our results reveal the first direct observations of acous- Imperial College London, London, UK tic streaming in a soft tissue microenvironment. We showed that al- Correspondence: Ahmed Elghamrawy though gelatin and polyacrylamide have acoustic properties suitable Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P42 for ultrasound imaging and hyperthermia experiments, they are not ef- fective in studying acoustic streaming. Instead, we recommend the use OBJECTIVES Acoustic streaming - the displacement of fluid by sound of MPPA, because it supports fluid flow through its microstructure by - has been proposed as the mechanism for therapeutic effects, such having interconnected pores or channels. Using this phantom material, as drug distribution enhancement and neurostimulation, yet there we wereable to study acoustic streaming under different exposure has been no direct observation or characterisation of this effect in parameters; finally demonstrating that pulsed ultrasound pro- soft tissue microenvironments, making it difficult to optimise and duces greater acoustic streaming with less risk ofthermal damage control. Post-mortem and indirect analyses have revealed changes in than with continuous wave emission using the same acoustic en- dye distribution and neuronal excitation, but it has remained uncer- ergies. Our future work is to study optimise ultrasound parame- tain whether these biological outcomes were due to acoustic stream- ters to maximise acousticstreaming in this material and in in-vivo ing or other responses, such as acoustic cavitation. We aimed to be tissue for enhancing drug distribution and neurostimulation. the first to directly observe ultrasound-induced streaming during sonication in a tissue-mimicking material. METHODS Our initial challenge was to create an appropriate material for studying acoustic streaming. We hypothesised that existing hydrogels (e.g., polyacrylamideand gelatin) used as ultrasound phan- toms, mimic the acoustic properties of tissue, but not the tissue microenvironment. We created three phantoms – gelatin, polyacryl- amide (PAA), and a new macroporous polyacrylamide (MPPA) hydrogel – and analysed their structure using scanning electron mi- croscopy (SEM) and mercury intrusion porosimetry (MIP). Based on our findings, we selected the MPPA for our tissue-mimicking material. Fig. 1 (abstract P42). Scanning electron microscopy of (a) gelatin, Focused ultrasound (FUS) pulses were emitted from asingle-element (b) polyacrylamide and (c) macroporous polyacrylamide. Zoomed in transducer [centre frequency (fc): 5 MHz], which was driven by a sections show closed pockets or pores for gelatin and PAA but, function generator through a power amplifier. The focal point of a MPAA pore structure provides fluid a path to follow throughout the FUS transducer’s beam (axial FWHM: 3.2 mm, lateral FWHM: 0.45 gel. Fluid can travel from the pore on the gel surface to the deeper mm) was placed at the distal surface of the MPPA. A model drug pore behind (bottom right rectangular areas) (Bromophenol blue) was injected in and around the focal region and Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 81 of 122 and synthesised multifunctional human serum albumin-chlorin e6 nanoassemblies (HSA-Ce6 NAs) that integrate imaging andtherapy functionalities into a single nano-platform for MRI guided sonody- namic therapy. We show the albumin-sonosensitizer nanoassemblies is highly accumulated in tumor tissue and had highly effective in destructing cancer cells under ultrasound and MRI-visible in vivo and thus, suitable for theranostic applications in cancer. METHODS HSA-Ce6 NAs synthesis as follows: after dissolution of HSA Fig. 2 (abstract P42). From images acquired, differences images at in water, Ce6 was added into the solution to precipitate the HAS- 1, 5 and 10 s after the start of the sonication (Pneg: 8MPa, PL: cotinuous Ce6, Then HSA-Ce6 NAs were mixed with MnCl2 solution at a molar wave) were obtained (a-c). The black circle denotes the sonicated ratio of 2:1 for the HSA-Ce6 NAs.To detection of ROS in HSA-Ce6 region, as the darker dye is cleared from hydrogel, the lighter clearer gel NAs, different samples were mixed with 1 mM 2′,7′-dichlorofluorescin surface is shown causing an increase in optical density diacetate (DCFH-DA), and then irradiated by different US times or US intensities.In vitro SDT: The ultrasound transducer (diameter: 1.5 cm, resonance frequency: 0.5 MHz, duty factor: 50 %, interval: 100 ms; ultrasonic intensity: 0.5 W/cm2) was fixed at the bottom of water bath. The cell culture plate was suspended 10 cm above the ultrasound transducer. In vivo MRI: The in vivo MRI imaging experiments were acquired using a 3.0 T clinical MR scanner (TIM TRIO, Siemens, Germany) with a small animal coil. T1-weighted MR images were acquired using the following parameters: TSE sequence, TR= 700 ms, TE =13 ms, FOV =32 × 45 mm, slice thickness= 1 mm and flip angle=180°. RESULTS The sonodynamic effect of HSA-Ce6 NAs was confirmed by measuring reactive oxygen species using DCFH-DA as a detector, as shown in Fig. 1, thefluorescence intensity of ROS exhibits a time- dependent and an intensity-dependent enhancement, indicating ROS from HSA-Ce6 NAs upon ultrasound irradiation. Thecell uptake behavior of HSA-Ce6 NAs was investigated through confocal micros- copy, the HSA-Ce6 NAs has high cell uptake efficiency with high fluorescence intensityin U87 cells (Fig. 2). Moreover, the SDT in vitro of HSA-Ce6 NAs could be further improved. It was found that single HSA-Ce6 NAs or ultrasound treatment could only induce partial cell death at the current conditions. In marked contrast, the combination treatments (SDT) were found to be highly effective in destructing cancer cells. Indicating that ultrasound has a good penetrability and SDT effect (Fig. 3). In addition, the HSA-Ce6 NAs can serve as chelat- Fig. 3 (abstract P42). Clearance rate is the average of optical ing agents to capture Mn2+ for MRI imaging by forming stable che- density values of the pixels in the black circle (Fig. 2) over time. As lates. The in vivo T1-weighted MR imaging of U87 tumor-bearing continuous wave ultrasound pressure increases from Pneg: 8MPa mice was carried out after iv injection of HSA-Ce6 NAs, The T1 signals (red) to 12MPa (green) there is a slight increase in dye clearance of tumor on mice strengthened with the increase of time interval, rate. Pulsed ultrasound was applied (Pneg: 25MPa, PL: 5 ms, PRF: 20 and reached a peak after 24 h postinjection of HSA-Ce6 NAs (Fig. 4). Hz) having thesame average intensity as 8MPa continuous wave This indicated the HSA-Ce6 NAs could provide the optimal time win- and an increase in clearance rate was observed (blue). All applied dow for sonodynamic therapy so that the ultrasound irradiation parameters initiated dye clearance compared to whenno ultrasound could be conducted in the targeted lesion. was applied (black) CONCLUSIONS We have shown our human serum albumin-chlorin e6 nanoassemblies are indeed visible by MRI in vivo and that they can be targeted by FUS to deliver and release reactive oxygen spe- cies (ROS) to kill cancer cells, paving the way for their theranostic ap- P43 plications under MRI-guided sonodynamic therapy. MRI-guided focused ultrasound mediated smart albumin- sonosensitizer nanoassemblies for sonodynamic therapy of glioma 1 1 2 1 2 Qian Wan , Dehong Hu , Mengjie Chen , Zonghai Sheng , Jun Zhou , 1 1 Xin Liu , Hairong Zheng Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences, Shenzhen, Guangdong, China; Medical College of Chinese Three Gorges University, Yichang, China Correspondence: Qian Wan Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P43 OBJECTIVES Sonodynamic therapy (SDT) represents an emerging ap- proach that offers the possibility of non-invasively eradicating solid tumors in a site-directed manner, which involves the synergistic ef- Fig. 1 (abstract P43). The ROS generation of HSA-Ce6 NAs under fect on cell damage by the combination of the sonosensitizer and US irradiation. (A) Fluorescence emissin spectra of HSA-Ce6 NAs in ultrasound, The sonosensitizer is the key factor of SDT.Chlorin DCFH-DA solution with the increase of US irradiation time. (B) e6(Ce6) has been widely used as a sonosensitizer in sonodynamic Fluorescence emissin spectra of HSA-Ce6 NAs in DCFH-DA solution therapy (SDT) on many human tumors, indicating Ce6 possesses ex- with the increase of US irradiation intensity cellent sono-activities with tiny toxicity.In this study, we designed Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 82 of 122 bony tissues. To devise ascheme for performing thermometry based on these MR properties, we first must know the relationship between these variables and temperature. The first objective of this investiga- tion was to quantify these relationships in ex-vivo tissue, with the eventual goal of developing a strategy for performing such measure- ments in-vivo during FUS sonication. To date, such measurements have been performed using 3D ultrashort echo time (UTE) acquisition pulse sequences, which severely limits the achievable temporal reso- Fig. 2 (abstract P43). Fluorescence images displayed cellular lution. Therefore, our second objective was to explore the feasibility localization of HSA-Ce6 NAs after 3 incubation with HSA-Ce6. Red of performing accurate temperature measurements in cortical bone represented the fluorescence HSA-Ce6NAs. Blue represented the using much faster 2D acquisitions with slice selective RF pulses. fluorescence of DAPI and green represented the fluorescence of the METHODS We measured the temperature dependence of T1 and cell membrane of U87 cells T2* in cortical bone from a recently euthanized swine tibia and bovine long bone using both 2D and 3D pulse sequences. In order to accurately measure these MR properties, we had to cre- ate a system in which we could vary the temperature of the bone as well as keep the temperature constant within the MR scanner during MR imaging. To achieve this, we built a system that consisted of an MR-safe water bath, a combination water- heater/recirculation pump that regulated the temperature of the water bath, and fiber-optic temperature sensors that monitored the temperature of a bone sample immersedin the water bath (Fig. 1). The bone sample was placed inside of the regulated water bath and its temperature was monitored by the fiber optic sensors while we acquired our measurements that would allow us to calculate the temperature dependence of the MR varia- bles.T1 was measured at 3T by acquiring a series of spoiled gra- dient-echo ultrashort UTE images, each having the same TR and Fig. 3 (abstract P43). The in vitro SDT in U87 cells. (A) Schematic TE but different flip angles, and then fittingthe measured signal- diagram of the insonation device. (B) Quantitative evaluation of cell versus-flip angle curve to the theoretical dependence at each survivals of each group voxel contained entirely within the bone. In order to find this theoretical dependence, we derived an equation that would give us the steady-state incoherent signal as a function of T1 and T2*. T2* was measured by acquiring a series of spoiled gradient-echo images, each having the same TR and flip angle but different echo times, and then fitting the measured signal-versus-TE curve to a mono-exponential decay at each voxel contained entirely within the cortical bone. The T1 and T2* measurements were per- formed at different temperatures ranging from below body temperature to nearly60°C. The temperature of the water bath was held constant during each T1 and T2* measurement. Fig. 4 (abstract P43). In vivo MRI images of the mice bearing U87 RESULTS T1 and T2* were both found to increase approximately linearly tumor after iv injection of HSACe6 NAs at different times. I.V. with temperature. The slope of this linear dependence was measured to injection dose of HSACe6 NAs is 2.0 mg Ce6/kg be 1.25 ms/°C for T1and 4.66 μs/°C for T2* in swine tibia and 1.38 ms/°C for T1 and 3.69 μs/°C for T2 in bovine long bone, utilizing the 3D imaging pulse sequence (Fig. 2). Additionally, we measured T1 and T2* in bovine bone using the 2D imaging pulse sequence, yielding slopes of 0.74 ms/°C P44 for T1 and 4.26 μs/°C for T2* (also Fig. 2). The quality of the theoretical fits Towards rapid MR thermometry in cortical bone 1 1 2 3 to the measured signal was generally excellent (Fig. 3). Phoebe Miller , Sina Tafti , David Keder , Quinton Miller , Darius 3 1 CONCLUSIONS The T1 variation with temperature was found to Hossainian , Wilson Miller be similar in swine and bovine cortical bone when measured Radiology and Medical Imaging, University of Virginia, Charlottesville, using a 3D pulse sequence with a nonselective excitation RF Virginia, USA; Physics, University of Virginia, Charlottesvle, Virginia, USA; pulse. However, the T1 values measured using a 2D pulse se- Biomedical Engineering, University of Virginia, Charlottesvle, Virginia, quence with a slice-selective excitation RF pulse were vastly dif- USA ferent. This discrepancy is almost certainly due to naive Correspondence: Phoebe Miller application of the theoretical signal versus T1 relationship, which Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P44 does not account for the non-ideal slice profile of our slice-select- ive excitation RF pulse. Such effects must be considered when OBJECTIVES Bone metastases are an important, FDA-approved treat- performing T1-based thermometry using 2D acquisitions. By con- ment target for MR-guided focused ultrasound therapy. However, it trast, the T2 variation with temperature was found to be similar is not currently possible to noninvasively measure temperature in when the same sample was measured using 2D and 3D acquisi- bony tissues. Conventional PRFS-based MR thermometry is likely not tions. Thus it may prove to be more straightforward to achieve feasible in cortical bone due to very fast T2* signal decay. Other MR quantitative accuracy when performing T2-based thermometry using properties, including T1 and T2*, also depend on temperature and 2D acquisitions. Whereas our 3D pulse sequence required 1 minute to thus offer alternative pathways for performing MR thermometry in Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 83 of 122 acquire a single image, our 2D pulse sequence required only 2.5 sec- OBJECTIVES In recent years, with the rapid development of medical onds. This time scale is much more favorable for monitoring imaging, ultrasound molecular imaging becomes one of the hot temperature changes during FUS application in vivo. spots in molecular imaging research field. METHODS The design of molecular probes is the key point and prerequisites for ultrasound molecular imaging. People increas- ingly pay more attention to the targeted ultrasound contrast agents which are the ultrasound molecular probes. And the inter- section of multiple disciplines will promote the development of ultrasound molecular imaging. RESULTS It is also urgent to develop a set of special equipment for efficient ultrasound molecular imaging. CONCLUSIONS The ultrasound molecular imaging instrument, micro- bubble/microsphere triggered device, imaging monitoring and ultra- sound molecular probes can be integrated into the low intensity ultrasound molecular imaging and therapy system, which will hope- fully bring about the integration of ultrasound molecular imaging, in vivo drug deliveryand controlled release and evaluation of treat- ment efficacy. It provides an innovative research platform for ultra- sound molecular imaging and therapy. Fig. 1 (abstract P44). Experimental apparatus P46 MR and fluorescence dual-modal imaging-guided sonodynamic therapy of gliomas through a multifunctional theranostic nanoplatform 1 2 1 Fei Yan , Meijun Zhou , Hairong Zheng Shenzhen Institutes of Advanced Technology, Shenzhen, China; Department of Ultrasonography, The Third Affiliated Hospital of Southern Medical University, Guangzhou, China Correspondence: Fei Yan Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P46 OBJECTIVES The aim of this study is to develop a multifunctional theragnostic nanoplatform for MR and fluorescence dual-modal im- aging-guided sonodynamic therapy of gliomas. METHODS Sinoporphyrin sodium (DVDMS), a NIR-absorbing sonosen- Fig. 2 (abstract P44). T1 and T2* dependence on temperature in sitizer and photosensitizer, was firstly used to chelate with manga- both swine and bovine bones when measured with 3D techniques, nese ion (Mn2+) and then encapsulated into liposomes by thin-film and in bovine bone when measured with 2D techniques rehydration method to fabricate DVDMS-Mn-Liposomes (DVDMS-Mn- LPs). The characterizations of DVDMS-Mn-LPs, their imaging capabil- ity in vitro and in vivo and SDT effect in subcutaneous and orthotopic glioma mouse models were examined (Fig. 1). RESULTS The resulting nanoparticles are proved to be physiologically stable and biocompatible, allowing time-dependent and intensity- dependent generation of oxygen free radicals upon ultrasound irradi- ation. Good T1-weighted MR and fluorescence imaging capabilities were demonstrated. We further employed this nanoparticle to treat subcutaneous and orthotopic glioma, demonstrating that SDT with DVDMS-Mn-LPs significantly improved the anticancer effect than that of PDT with DVDMS-Mn-LPsin the presence of skull. Histological ana- lysis further revealed much more apoptotic cells and lower tumor cell proliferation, confirming the advantageous anti-tumor effect of SDT over PDT against glioma. CONCLUSIONS Our study developed a novel theragnostic nanoplat- Fig. 3 (abstract P44). T1 and T2* measurement data, together with form and provided a promising strategy for imaging-guided SDT for best fit curves, for swine tibia glioma treatment. P45 Multi-disciplinary integration of ultrasound molecular imaging Zhigang Wang Institute of Ultrasound Imaging, Chongqing Medical University, Chongqing, China Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P45 Fig. 1 (abstract P46). See text for description Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 84 of 122 P47 OBJECTIVES Biofilm's control is a critical issue for water treatment Comparative study on the characteristics of PLGA microspheres systems. The primary goal is to use low-intensity pulsed ultrasound with different drugs to promote liposomes which encapsulate antibiotics into agarose- Dong Yu film based bio-film phantoms and release the encapsulated antibi- Chongqing Medical University, Chongqing, China otics by applying mild-intensity focused ultrasound. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P47 METHODS Liposomes were synthesized using 1,2-Diacyl-sn-Glycero-3- Phosphocholine (PC) and 1,2-Dipalmitoyl-sn-Glycero-3-Phosphoethanola- OBJECTIVE To compare the characteristics of PLGA microspheres mine (DPPE) mixedto a percent molar ratio of 95:5 and dissolved in containing different drugs and green fluorescent dextran, to provide chloroform in a round-bottomed flask and then were dialyzed against the basis for drug treatment of fungi. PBS for 24 hours. Dialysate (one liter) was changed 4-6 times, resulting in METHODS PLGA microspheres were prepared by double emulsification a relatively pure liposome suspension that encapsulate a gentamicin so- method, and different antifungal drugs were encapsulated, as well as green lution, Liposome size distributions were determined by the autocorrel- fluorescent dextran. The experiment consisted of PLGA microspheres, PLGA ation function of the dynamic light scattering method. The1.5% sodium microspheres containing green fluorescent dextran, PLGA microspheres alginate solution was prepared by adding 3g alginic acid, sodium salt loaded with amphotericin B and PLGA microspheres loaded with flucona- and 0.6mL glycerol to 200 ml distilled water in a 250ml beaker via con- zole. The surface morphology, internal structure, particle size, potential, en- tinuous vortex-mixing overnight. Biofilms were generated using 1x106 trapment efficiency and drug loading were measured and compared. Ralstonia insidiosa in 1.5% alginate solution and allowed to stand for 5 RESULTS (1) The PLGA microspheres were uniform in size, well dis- minutes to assure that the surface of the solution flat. 600 ml of 2% CaCl2 persed and spherical in shape and regular in morphology. (2) The was added to the alginate from the top for two hours. After particle size of PLGAmicrospheres was 343.17 ± 44.5nm. (3) PLGA polymerization, the film was washed with sterile water and incubated microspheres containing green fluorescent dextran were used as the with R2B overnight at room temperature. A two-step treatment was ap- experimental control group, and the green fluorescence was ob- plied. The first step is the penetration of the liposomes into the biofilm. served under laser confocal microscope. Indicating that PLGA micro- The front part of the transducer was immersed in the liposome suspen- spheres successfully encapsulated green fluorescent dextran. sion. The distance between the transducer surface and the top of the al- CONCLUSIONS PLGA microspheres, PLGA microspheres loaded with ginate-flm was 1cm.Tone-bursts ultrasound (2.25MHz with duty cycle PLGA and PLGA microspheres have good physical properties under 10%) were emitted downward into the liposome solution. Experiments the same conditions, which lays a foundation for the synergetic ef- were conducted with ultrasound spatially and temporally averaged inten- fect of microbubbles in the treatment of fungi. sity, ISATA = 0.14 W/cm2.The ultrasound transducer used is a single-elem- ent non-focusing piezo-ceramic transducer operated at 2.25 MHz, with active radius a =10 mm. An arbitrary waveform function generator was P48 programmed to produce a tone-burst sinusoidal signal of duty cycle of Therapeutic effect of focused ultrasound combined with anticancer 10% for 1 min insonation duration, and the output of the waveform gen- drug loaded microbubble complex for pancreatic cancer: erator was used as the input of a 55 dB RF power amplifier whose output preliminary study was used to drive the transducer. The experiment step two uses a fo- Eun-Joo Park, Yun Deok Ahn, Yuri Cheon, Jae Young Lee cused ultrasound to burst the liposomes inside the biofilm and release Radiology, Seoul National University Hospital, Seoul, Korea the drug from the liposomes in situ. The sample petri dish was mounted Correspondence: Eun-Joo Park and submerged in a tank containing filtered, distilled and degassed Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P48 water, the position of the sample petri dish can be transported in three orthogonal dimensions by a computer-controlled manipulator with step OBJECTIVES As focused ultrasound (FUS) combined with microbub- distance of 1 mm. A piezo-ceramic transducer operating at f0 =1.1 MHz bles has been widely studied in cancer treatment, there is growing (geometric focal length d = 62 mm, active radius, a = 32 mm) was interests in developing nanoparticles for FUS enhanced cancer drug mounted at the bottom of a tank filled with degassed water facing up- delivery. This study was designed to evaluate therapeutic effects of ward. The distance between the transducer and the biofilm wasadjusted anticancer drug loaded microbubble complex in combination with to be equal to the geometric focal length (d=62mm). The same signal focused ultrasound treatment for pancreatic cancer. generator and amplifier were used, ultrasound tone-bursts (1.1MHz with METHODS Immunodeficient mouse inoculated with CFPAC-1 were duty cycle 10%) were emitted upward to the biofilm. The sample was used as the pancreatic xenograft model for in vivo studies. Animals moved in a horizontal plane near the focal plane of the source transducer were treated in five groups: control, Doxorubicin-only (Dox), Doxo- by a manipulator, the ultrasound focal point was scanned all over the rubicin combined with FUS treatment (Dox-FUS), Doxorubicin loaded biofilm with 1 mm gap. The spatially- and temporally-averaged acoustic microbubble complex (MB-NP-Dox) only, and MB-NPDox combined power was measured using the radiation force method to be 60 W/ with FUS treatment (MB-NP-Dox-FUS). Animals were treated on a cm2by the measured acoustic power divided the surface area of the weekly basis for three weeks and post-treatment monitoring was transducer. Experiments on a control group were conducted, following followed for five weeks. the same procedure but keeping the ultrasound off. Additional control RESULTS As results, therapeutic effects of MB-NP-Dox-FUS will be was done by repeating the experiments using gentamicin solution alone presented as well as the side effects of each treatment to address without liposomes. All treatments were repeated at least three times. the safety of using MB-NP-Dox. RESULTS Bacterial colony forming units were measured. The four dif- CONCLUSIONS Based on this result, further study will be followed to ferent cases include: antibiotic alone shown in Fig. 1 A; ultrasound improve therapeutic effects of the combined treatment as well as alone US shown in Fig. 1 B; ultrasound plus the antibiotic drug in so- the drug loading efficiency of MB complex. lution shown in Fig. 1 C; and focused ultrasound and the antibiotic encapsulated in liposomes shown in Fig. 1 D. CONCLUSIONS Our experimental results have shown that the fo- P49 cused ultrasound can burst the liposomes and release the drugs. Its killing effects were very close to the case that uses the detergent to Bio-film mitigated by drug loaded liposomes promoted by pulsed lyse the liposomes. The focused ultrasound plus gentamicin-encapsu- ultrasound 1 2 Junru Wu , Faqi Li lated liposomes (D) reduce the number of viable bacteria residing 1 2 University of Vermont, Burlington, Vermont, USA; Chongqing Medical inalginate based biofilm by 72 percent (p<0.001). Acknowledgment: University, Chongqing, China Funded by NASA (Cooperative agreement number NNX13AD40A) & Correspondence: Junru Wu National Natural Science Foundation of China (Grant Nos. 81127901, Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P49 11574039,81201102, 11274404). Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 85 of 122 P51 Multi-stage surface acoustic wave for separation of cancer cells in microfluidic device Kaiyue Wang, Wei Zhou, lili niu, Feiyan Cai, Fei Li, Long Meng Shenzhen Institutes of Advanced Technology, Shenzhen, China Correspondence: Kaiyue Wang Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P51 OBJECTIVES The separating of circulating tumor cells (CTCs) is significant in the early diagnosis, prognostic judgement, and provides an effective evidences for guideness of clinical chemotherapy. In order to detect CTCs effectively, many separation techniques have been developed to date, in- cluding functionalized polymers, pinched flow fractionation, hydro- dynamic filtration, inertial microfluidics, deterministic lateral displacement, fluorescent activated cell sorting. With the advantages of non-contact and high throughput, acoustic manipulation has received increasing attention in the cell separation. To ensure the focused cells at a certain position relative to the microchannel, it is essential to align the piezoceramic trans- ducer (PZT) and microchannel precisely, which affects the stability and re- liability of the device. In this paper, a microfluidic device with two Fig. 1 (abstract P49). A statistical analysis was performed using the acoustic stages has been developed to concentrate and separate the can- pairwise comparison “FTest Two Samples for Variances” method. The cer cells without the need of the precise alignment and sheath flow. number of viable bacteria in the alginatebased biofilms were METHODS The standing surface acoustic wave (SSAW) device including a reduced by 72 percent. No statistical difference (p>0.05) was pair of straight interdigital transducers (IDTs) and a pair of circular IDTs found among A, B, C cases was fabricated on the surface of a 128°<span style="font-size:12px; line- height:19.2px"> </span>Y-rotated, X-propagating LiNbO3 substrate. The straight IDTs were used to generate SSAW for cell concentration and the advantages of circular transducers were exploited to generate TSAW for cell separation. The polydimethylsiloxane (PDMS) microchannel was bonded to the SSAW generator using oxygen plasma treatment. And a sample of the mixture of U87 cancer cells and RBCs was used to demon- P50 strate the manipulations of concentration and separation. The frequencies Preparation and characteristics of targeted phase-shift lipid of continuous signals and pulse signal sent to the straight IDTs and the fo- nanoparticles mediated by tumor homing and penetrating peptide cused IDTs were set to be29.74 MHz and 38.74 MHz, respectively (Fig. 1). Leilei Zhu RESULTS We examined the ability of the device by manipulating micro- Ultrasound Imaging, Chongqing, China spheres with about 2μm diameter. When the stream flowed in SSAWs, the Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P50 microspheres promptly arranged in the center of the microchannel. By modulating the relative phase, △Φ, between two IDTs, the position of the OBJECTIVES To prepare a novel ultrasound contrast agent, microspheres relative to the microchannel could be adjusted arbitrarily. targeted phase-shift lipid nanoparticles mediated by tumor When the concentrated cells passed through TSAWs, the microspheres homing and penetrating peptide tLyP-1, and to evaluate its were accurately separate into two categories under theaction of radiation characteristics. force generated by TSAWs, and thus the microspheres separation could METHODS The nanoparticles were prepared by filming-rehydration be achieved. and acoustic-vibration methods. The morphology, distribution, par- CONCLUSIONS The experimental results reveal that the multi-stage ticle size and zeta potential were detected. After heating and irradiat- acoustic-based approach utilizing microfluidic device a promising tech- ing of low intensity focused ultrasound (LIFU), the phase-shift nique in effectively realizing the concentration and separation of cells. characteristic and the enhancement effect in vitro were observed. This kind of acoustic-based devices can avoid the cellular damage caused The tumor homing and cell-penetrating properties of the nanoparti- by shear force which is generated by water dynamics. In addition, the cles were examined by confocal laser scanning microscopy and flow use of this method can avoid precise alignment of microchannel and cytometry. The cytotoxicity of the nanoparticles was evaluated by IDTs, improving the stability and practicability of the system. CCK8 assay. RESULTS Thesizeand distribution of nanoparticles were uni- formed. The size and zeta potential of nanoparticles were (399.50±29.98) nm and (3.28±1.72) mv, respectively. When the nanoparticles were heated to a temperature of 45 °C or after ir- radiated by LIFU, nanoparticles generated phase-shift and en- hanced ultrasound imaging in vitro(P<0.05). The confocal laser scanning microscopy showed that nanoparticles can targetedly aggregate to cell membrane of MDA-MB-231 and penetrate into the cell, but not to HUVEC. The flow cytometry showed that intracellular fluorescence intensity of MDA-MB-231 was higher than that of HUVEC(P<0.05). The CCK8 assay indicated that dif- ferent concentrations of nanoparticles had no significant effects on cell activity (P>0.05). CONCLUSIONS A novel ultrasound contrast agent, targeted phase-shift lipid nanoparticles mediated by tumor homing pene- trating peptide tLyP-1, was prepared successfully. It can target to MDA-MB-231 cell and penetrate into the cell in vitro,and en- Fig. 1 (abstract P51). Schematic diagram of experiment. 1. The hance ultrasound imaging in vitro after LIFU irradiation, which microspheres injected into the microchannel. 2. The microspheres expected to be a novel tumor targeted ultrasound contrast align in the center of the microchannel by the SSAW 3. The microspheres agent and achieve ultrasound molecular imaging at the level of were driven to the upper wall of the channel by the focused TSAW tumor cell. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 86 of 122 P52 Magnetic nanoliposomes as in situ microbubble bombers for multimodality image-guided cancer theranostics Yang Liu, Fang Yang, Chuxiao Yuan, Mingxi Li, Tuantuan Wang, Ning Gu School of Biological Science and Medical Engineering, State Key Laboratory of Bioelectronics, Jiangsu Key Laboratory for Biomaterials and Devices, Nanjing, JiangSu, China Correspondence: Yang Liu Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P52 Fig. 2 (abstract P52). Fabrication and characterization of AMLs and size distribution of liposomes, ALs, and AMLs, respectively OBJECTIVES The physiological barriers imposed by the abnormal tumor vasculature and the dense collagen matrix have prevented nanocarriers from being delivered to unperfused deep tumor regions, which results in weakening the effectiveness of cancer therapeutics. Herein, we designed a nanoliposome drug delivery system aimed at the improvement of nanocarrier accumulation and distribution in the tumor’s poorly accessible regions for excellent diagnosis and potential therapeutic effects for imaging-monitored accurate tumor ablation. METHODS The anethole dithiolethione (ADT) loaded magnetic nano- liposome (AML) delivery system was prepared by a hydration and membrane-filtering method, which consists of ADT, hydrogen sulfide (H2S) pro-drug, doped in the lipid bilayer, and superparamagnetic nanoparticles (MNPs) encapsulated inside (Fig. 1). The size distribu- tion and morphology of the liposomes were characterized (Fig. 2). The generation of H2S gas in HepG2 cells was investigated by a real- time live cell optical imaging system (Figs. 3, 4). High-resolution MRI (7T) and microbubble-enhanced US imaging was performed pre- and post-injection of AML. The therapeutic efficiency of the AML in the tumor-bearing nude mouse model was also evaluated. RESULTS The presence of MNPs trapped and dispersed in the core of the liposome (about 200 nm) was confirmed. The optical microscopic images of a Hep G2 cell showed when incubated with AMLs, a change in cellular morphology and even cell rupture was observed with the in- crease of incubation time. After 6h of incubation, the cells were dis- rupted and detached from the dish bottom. For in vivo applications, when intravenous injection of AMLs, the targeting of AMLs to tumor was enhanced under exposure to an external magnetic field. Time- dependent in vivo MR and US imaging of tumors in a HepG2-bearing mouse model was significantly enhanced. Moreover, after 7-day follow- up observation, AMLs with magnetic field treatments have indicated extremely significantly higher inhibitions of tumor growth. CONCLUSIONS In summary, AMLs are feasible as both synergistic Fig. 3 (abstract P52). Live cell optical system for observation of cell agents to strengthen tumor ablation efficiency and dual-mode con- morphology change after incubation with liposomes, ALs, and AMLs. trast agents to provide significant contrast enhancement for MR and For each sample, a time gradient was acquired ultrasound imaging. This proposed strategy with both enhanced tumor accumulations of liposomes as well as the property of nano- particles to microbubbles conversion holds great promise for multi- modal image-guided accurate cancer therapy. Fig. 4 (abstract P52). Timedependent in vivo MR, US and NIRfluorescence imaging of tumors after intravenous injection of Fig. 1 (abstract P52). Concepts and schematics of AMLs and their samples in a HepG2bearing mouse model, and histological analysis nano to micro conversion for US/MR dual modal imaging and the of excised tumors and major organs after AMLs with and without spatiotemporal bombed combination tumoraccurate therapy magnetic field (MF) treatment Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 87 of 122 P53 OBJECTIVES To prepare a hematoporphyrin monomethyl ether (HMME)- Incubation of free bubble water and phospholipids to prepare loaded poly (lactic-co-glycolic acid) (PLGA) microcapsules (MBHMME/ shelled nanobubbles ultrasound contrast agents PLGA), which could not only function as efficient contrast agent for ultra- Jilai Tian, Fang Yang, Juan Jin, Ning Gu sound (US)/photoacoustic (PA) imaging, but also as a synergistic agent Biological Science and Medical Engineering, Southeast University, for high intensity focused ultrasound (HIFU) ablation for its sonodynamic Nanjing, Jiangsu, China therapy(SDT). Correspondence: Jilai Tian METHODS MBHMME/PLGA was prepared with the double emulsion Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P53 evaporation method by Sonosensitizer HMME nanoparticles were in- tegrated into PLGA microcapsules. After characterization, the cell-kill- OBJECTIVES Nanobubbles less than 1 μm could make a promising ing and cell proliferation-inhibiting effects of MBHMME/PLGA application in ultrasound molecular imaging and drug delivery. How- microcapsules on ovarian cancer SKOV3 cells were assessed. The ever, the fabrication of stable gas encapsulation nanobubbles is still characteristic fluorescence peak of FCLA free acid (reactive oxygen challenging. Free gas bubbles without any shell materials may have species (ROS)-based chemiluminescence reagent) at around 520 nm lots of applications such as water purification, drag reduction, gener- was detected with the fluorescence spectrophotometer. The US/PA ation of free radicals and so on. Herein, we try to make lipid shelled imaging-enhancing effects and synergistic effects on HIFU were eval- nanobubbles with free gas bubble water. uated both in vitro and in vivo. METHODS Phospholipids dried under vacuum or the lyophilized RESULTS MBHMME/PLGA were highly dispersed with well-defined phospholipids were employed. The characters of size, size distribu- spherical morphology (462 ± 0.52 nm in diameter, PDI = 0.932). En- tion, surface tension, viscosity and their ultrasound imaging in vitro capsulation efficiency and drug loading efficiency were 58.33 ± of the prepared lipids-shelled nanobubbles were investigated. The 0.95% and 4.73 ± 0.15%, respectively. The MBHMME/PLGA remark- mechanism of the incubation was also discussed. ably killed the SKOV3 cells and inhibited the cell proliferation, signifi- RESULTS Results show that this type of GU-Liposome has mean cantly enhanced the US/PA imaging results and greatly enhanced diameter of 194.4± 6.6 nm and zeta potential of -25.2± 1.9 mV with the HIFU ablation effects on ovarian cancer in nude mice by the layer by layer self-assembled lipid structure. The acoustic imaging HMME-mediated sono-dynamic chemistry therapy (SDT). analysis in vitro indicated that ultrasound imaging enhancement CONCLUSIONS MBHMME/PLGA represents a potential multifunc- could be acquired by both perfusion imaging and accumulation im- tional contrast agent for tumor diagnosis and treatment, which aging. The dispersed phospholipid molecular in the prefabricated might provide a novel strategy for thehighly efficient imaging-guided free nanobubbles water was expected to be assembled to form con- non-invasive HIFU synergistic therapy for cancers by SDT in clinic. trollable stable lipid encapsulation gas containing ultrasound-sensi- The mechanism of singlet oxygen could be generated from tive liposome (GU-Liposome). Compared with conventional HMMEand MBHMME/PLGA irradiated with HIFU needed further in- mechanical agitation methods, pre-prepared free gas bubble-based vestigation in the future. nanobubbleshave exhibited the controllable nano-size, lower polydis- persity index. CONCLUSIONS Thus, by incubation of free bubble water and phos- P56 pholipids, stable lipid-shelled nanobubbles could be prepared to Preparation of phase-transition perfluoropentane nanodroplets broaden the biomedical application of nanobubbles in the theranos- modified by folate for ultrasound molecular imaging in ovarian tics in future. cancer Jianxin Liu Chongqing Medical University, Chongqing, China P54 Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P56 Compare ultrasound-mediated cavitation between flowing phase- shift nanodroplets and lipid-shelled microbubbles during focused OBJECTIVES This study aimed to develop a hybrid platform based on ultrasound exposures folate-modified phospholipid-shell and perfluoropentane nanodro- Siyuan Zhang, Tianqi Xu, Zhiwei Cui, Sihao Liu, Dapeng Li, Rui Guo, plets (FA-NDs), which could in vitro and in vivo target ovarian cancer Junjie Wang, Yujin Zong, Mingxi Wan and enhance ultrasound imaging after acoustic droplet vaporization Department of Biomedical Engineering, Xi'an Jiaotong University, Xi'an, (ADV) induced by low-intensity focused ultrasound (LIFU). Shaanxi, China METHODS The nanodroplet was fabricated with HSPC, DSPE-PEG Correspondence: Siyuan Zhang (2000) folate, DPPG, cholesterol and perfluoropentane using lipid film Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P54 hydration method and rotary evaporation method. The nanodroplet stability was evaluated at 4 and 37 respectively. The in vitro targeted ef- This abstract is not included as it has already been published: ficiency were tested with SKOV3 cells and in vitroADV was appraised in Zhang S, Cui Z, Xu T, Liu P, Li D, Shang S, Xu R, Zong Y, Niu G, Wang jellium model with LIFU. The in vivo targeted efficiency and acoustic S, He X, Wan M. Inverse effects of flowing phase-shift nanodroplets droplet vaporization were evaluated with SKOV3 tumor-bearing nude and lipid-shelled microbubbles on subsequent cavitation during fo- mice. cused ultrasound exposures. Ultrason Sonochem. 2017; 34: 400-409. RESULTS The nanodroplets were successfully prepared with good size Available from: http://www.sciencedirect.com/science/article/pii/ uniformity (particle size 321±67 nm). The nanoparticles remained stable S1350417716302139. for 48 h at 4 and 1 h at37. In vitro targeted experiments exhibited a perfect binding efficiency of FA-NDs to SKOV3 cells. In vitro ADV pro- files displayed obvious ultrasound enhancement in both B-mode and P55 CEUS-mode when LIFU power was elevated to 5 W. In vivo and ex vivo Sonodynamic Therapy (SDT) polymer contrast agent for fluorescence imaging displayed that FA-NDs possessed outstanding ultrasound/photoacoustic dual-modality imaging-guided specificity to targeted solid tumor. Both the qualitative and quantitative synergistic High Intensity Focused Ultrasound (HIFU) therapy results of in vivo ADV in mice tumor nodule manifested that FA-NDs Lan Hao underwent phase transition upon the LIFU exposure. Chongqing Key Laboratory of Ultrasound Molecular Imaging, the CONCLUSIONS The results demonstrated that the FA-NDs system is a Second Affiliated Hospital University, Chongqing, China potential platform for targeted conjunction and enhancing ultra- Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P55 sound imaging via ADV using LIFU. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 88 of 122 P57 ited good photostability. Upon near-infrared laser irradiation, the PLGA- Targeted Pegylated PLGA coated prussian blue nanocomposite for PB-PTX-PEG-FA nanocomposite showed an enhanced synergistic photo- dual-modality PA/MR imaging and synergistic chemo-thermal thermal and chemical therapeutic efficacy forbreast cancer than solo tumour therapy photothermal therapy or chemotherapy. Tingting Shang CONCLUSIONS In conclusion, we prepared highly dispersed core- Institute of Ultrasound Imaging Department of Ultrasound, The Second shell structure PLGA-PB-PTX-PEG-FA nanocomposite by coating PB Affiliated Hospital of Chongqing Medical University; Chongqing Key NPs with PLGA-FA shell and then modifying with PEG. The prepared Laboratory of Ultrasound Molecular Imaging, Chongqing, China. PLGA-PB-PTX-PEG-FA nanocomposite has good biocompatibility, ex- Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P57 cellent photo-thermal transformation capacity, and can enhance both PA imaging and MR imaging in vitro and in vivo. Furthermore, PLGA- OBJECTIVES To prepare a PEGylated poly (lactic-co-glycolic acid) PB-PTX-PEG-FA nanocomposite acts as a multifunctional drug deliv- (PLGA) targeting with folic acid (FA) coated Prussian blue nanoparti- ery system with higher drug loading capacity and excellent drug cles (PB NPs) and paclitaxel(PTX), to construct multifunctional PLGA- controlled release-system. Finally, our result confirmed that PLGA-PB- PB-PTX-PEG-FA nanocomposite for both photoacoustic (PA) imaging, PTX-PEG-FA nanocomposite has the ability of the synergistic thera- magnetic resonance imaging (MRI) and synergistic chemo-thermal peutic efficacy combined photothermal and chemical therapy effect, tumor therapy. which further enhances the therapeutic efficacy to breast cancer. METHODS Paclitaxel (PTX)-loaded Folic acid (FA) targeted PEGylated PLGA nanoparticles encapsulating Prussian Blue (PB) (PLGA-PB-PTX-PEG- FA) nanocomposite was fabricated by a modified double emulsion (water/oil/water) evaporation process. The morphology, size, UV–vis–NIR P58 absorbance spectra, Fouriertransfer infrared (FTIR) spectrum were tested Low intensity focused ultrasound regulates drug release from to evaluate the structural characterization of PLGA-PB-PTX-PEG-FA. Drug porphyrin-phospholipid liposomes and facilitates multi-functional Loading and releasing of PLGA-PB-PTXPEG-FA nanocomposite were theranostics 1,2 1, 2 1 1 assessed by high performance liquid chromatography (HPLC). Addition- Xiaobing Wang , Xiufang Liu , Fei Yan , Hairong Zheng ally, in vitro cell targeting and in vivo tumor targeting were verified by Shenzhen Institutes of Advanced Technology, Chinese academy of confocal laser scanning microscopy (CLSM) and vital fluorescence im- Sciences, Shenzhen, China; Shaanxi Normal University, Xi'an, China aging (VFI) to assess the targeting effect of PLGA-PB-PTX-PEG-FA nano- Correspondence: Xiaobing Wang composite. The efficacy of the contrast agent for PA imaging and MRI Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P58 was evaluated by in vitro and in vivo imaging of subcutaneous MDA-MB- 231 tumor-bearing mice following tail intravenous injection of the con- OBJECTIVES Nanoscale drug delivery systems (DDS) facilitate multifunc- trast agent. Finally, to visualize the photothermal cytotoxicity of PLGA- tional theranostics. Externally controlled drug release from DDS provides PB-PTX-PEG-FA nanocomposite, MDA-MB-231 cells were incubated with selective targeting. Here, we introduce an ultrasound-activatable porphy- PLGA-PB-PTX-PEG-FA nanocomposite in 96 well-plate for 2 h and were rin-phospholipid liposome (PPL) to show its bimodal imaging, robust car- irradiated with or without NIR laser. CCK8 method was used to measure rier, drug release behavior and efficient therapeutic actions. the viability of the cells. To study the photothermal effect and synergistic METHODS Dynamic light scattering and transmission electron micros- therapeutic efficacy, PLGA-PB-PTX-PEG-FA nanocomposite suspensions copy were utilized to characterize the features of PPL and PPL loaded of various concentrations nanoparticle dispersions were exposed to a with doxorubicin (PPLDOX) (Fig. 1). The ultrasound controllable drug re- laser with the wavelength of 808 nm and the output power of 0.647W lease was measured at distinct time points with or without different for 10min in vitro or in vivo after vein intravenous injection. free radical inhibitors (Fig. 2). In vitro cellular uptake and cytotoxicity RESULTS The highly dispersed PLGA-PB-PTX-PEG-FA nanocomposite were examined in U87 glioma cells by confocal microscopy and micro- with spherical morphology with smooth surface was directly observed by plate reader. In vivo, photoacoustic and NIR fluorescent imaging were TEM and SEM and has good uniformity with an average hydrodynamic applied to indicate the distribution of PPL following intravenous injec- diameter of 236.6±55.04nm. That PB NPs being encapsulated by PLGA- tion, along with the signals change after ultrasonic irradiation (Fig. 3). FA-PEG was confirmed by TEM, UV–vis–NIRabsorbance spectra and FTIR The tumor inhibition and overall survival timewere evaluated by PPL- spectrum of the PLGA-PB-PTX-PEG-FA and PB NPs. Both PB and PLGA- DOX endowed with the optimal porphyrin-phospholipid ratio, max- PB-PTX-PEG-FA nanocomposite displayed a broad absorption band from imum DOX loading and favorable ultrasound-responsible ability (Fig. 4). 500 nm to 900 nm with a strong absorption peak at ~702 nm. While the RESULTS The obtained PPL, PPL-DOX were stable at 4°C for at least two IR spectrum of the PLGA-PB-PTX-PEG-FA nanocomposite showed both months with appropriate 100 nm diameter. The ultrasound induced drug the corresponded peaks of PLGA-FA-PEG and PB NPs. The results of drug release from PPLDOX increased with higher molar phorphyrin-phospho- loading efficiency and drug loading capacity tested by HPLC were lipid, while the drug loading efficient declined as phophyrin-phospholipid 77.82% and 7.22%. The fast drug release from the nanocomposite in vitro ratio increased. 2 molar % porphyrin-phospholipid in PPL-DOX was com- with laser irradiation indicated the great potential as a controlled-release patible with good ultrasound-responsible and drug-loading capacities. system for the anticancer drug. In vitro cell targeting and in vivo tumor Under this condition, DOX release from PPL-DOX could beregulated by targeting of PLGA-PB-PTX-PEG-FA nanocomposite, targeted group had ultrasound intensity and abolished with different free radical inhibitors. Fol- obvious difference from non-targeted group and blank control group, lowing intravenous administration, the PPL demonstrates high sensitive which showed that good targeting effect to tumor cells or tumor in vivo photoacoustic and NIR fluorescent imaging and effective tumor site drug of PLGA-PB-PTX-PEG-FA nanocomposite. After the tail intravenous injec- delivery in xenograft U87-bearing mice. Ultrasound exposure triggers sono- tion of thePLGA-PB-PTX-PEG-FA nanocomposite, the PA images and MR dynamic damage of tumors cells and simultaneously initiates local drug re- images of tumor of nude mice were significantly enhanced, while there lease to exert chemotherapy. Exposure to focused ultrasound with PPL were almost no distinct enhancement in non-targeted group and blank suppresses tumor growth several times more than without exposure to control group. The result of the photothermal cytotoxicity of PLGA-PB- ultrasound. What’s more, PPL-DOX displays little damage on normal cells PTX-PEG-FA nanocomposite showed that the viability of the cells of in vitro and even on normal tissues in vivo. PLGA-PB-PTX-PEG-FA nanocomposite+Laser group was lowest. The re- CONCLUSIONS The developed ultrasound-activable paradigm sults of photothermal conversion property of PLGA-PB-PTX-PEG-FA sug- achieves simultaneous photoacoustic/fluorescence imaging and gested the character of photothermal effect was positively correlated spatiotemporally regulates nano-drug release and initiates sono- with the heating power and the nanoparticles concentration and exhib- chemotherapeutic effects. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 89 of 122 Fig. 1 (abstract P58). PPL-DOX characteristics. (a) Transmission electron microscopy of PPL-DOX liposome morphology under Fig. 4 (abstract P58). In vivo NIR flourescent imaging. (a) The different magnifications with 2% Pyro-lipid (Molar ratio). distribution of drug in tumor after I.V. injection. (b) Ultrasound (b) Particel size analysis of PPL-DOX with different Mol % comnbined with microbubbles enhanced the fluorescence of porphyrin-phospholipid Pyro-liposome, facilitating NIR and photoacoustic imaging and sonochemical therapeutics. A dual tumor model was used, with a tumor on the right flank was irradiated and the other was non-irradiated P59 Protective effects of static magnetic field on the sonoporation-induced cellular ionic imbalance Yaxin Hu, Mei Yang, Yancheng Wang, Siping Chen Biomedical Engineering, Shenzhen University, Shenzhen, Guangdong, China Correspondence: Yaxin Hu Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P59 OBJECTIVES During the past two decades, great efforts have been made to investigate the perforation and recovery dynamics of sono- poration-created pores on the plasma membrane (PM). Although both Fig. 2 (abstract P58). Ultrasound triggered drug release from acoustic and non-acoustic parameters have been extensively optimized PPL-DOX with (a) or without (b) free radicals scavengers to facilitate the pore-resealing process, sonoporation mediated intracel- lular delivery is still challenged by the fact that a significant proportion of the sonoporated cells would undergo apoptosis. It has been hypoth- esized in previous investigations that excessive ion exchanges across the PM pores could result in a cellular ionic imbalance, which might consequently activate the apoptosis signaling pathway. To provide dir- ect evidence for this hypothesis, this work will study the ion exchange dynamics in sonoporation and its potential relevance to apoptosis in- duction. Particularly, our work will also examine the question of whether the ion exchange dynamics in sonoporation could be modu- lated by a magnetic field. METHODS To investigate the ion exchange dynamics across the PM pores, a real-time sonoporation experiment platform was established. More specifically, an adherent microbubble (Targestar® P) was firstly intro- duced to the apical PM of a single MRC-5 cell. Then, inertial cavitation of the microbubble was triggered by a single ultrasound pulse (center fre- quency: 1 MHz; peak negative pressure: 0.85 MPa; pulse duration: 10 μs) and monitored by a high-speed camera (imaging rate of 680,000fps). In this way, the ion exchange dynamics in sonoporation could be imaged by a high-resolution confocal microscope (LSM 710). To record the cal- cium influx insonoporation, a green-fluorescent dye (Fluo 4, AM) was in- troduced to the cytoplasm. To evaluate the mitochondria damage in sonoporation, a red-fluorescent dye(TMRE) was used to indicate the mito- chondria membrane potential (MMP). Nucleic acid stains of Sytox Blue Fig. 3 (abstract P58). The characterization of PPL flourescene. and PI were added to the sonoporated cells at different timepoints (0 s (a) Flourescene emission of intact quenched PPL (in PBS) versus and 10 min post-sonoporation, respectively) to examine the efficiency of disrupted unquenched PPL. (b) Flourescene intensity of PPL upon PM resealing with and without a static magnetic field (680 mT). Qualita- ultrasound irradiation range from 0.25 Mpa to 0.35 Mpa. (c) Flourescene tive andquantitative analyses of fluorescent images were carried out photographs of PPL post ultrasound irradiation was recorded using the ImageJ software. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 90 of 122 RESULTS As illustrated by the green-fluorescent images in Fig. 1, cal- biofilms. This part of the study was conducted according to the method cium influxes (indicated by red arrows, top row) into the sonopo- described by Singh et al. rated cell (cell #2) could be found at the microbubble-adherent sites RESULTS Bacterial morphology. In group C, bacteria within biofilms (2 s post-sonoporation). The intracellular calcium concentration of were closely linked and wrapped by abundant fiber-like materials, caus- the sonoporated cells increased to the peak value at 4 s post sono- ing the formation ofhighly organized multicellular population structure poration, and the mitochondria membrane potential greatly changes (Fig. 1A). In group U, however, less fiber-like material was observed in at this time point (red fluorescence, bottom row). In the presence of biofilms structure. Some bacterial debris, however, was visible and the static magnetic field, both the ion influx speed and the MMP change bacterial surface exhibited deformation (Fig. 1B and C). Bacterial viability. amplitude were reduced. With respect to the PM resealing, the mag- Bacterial plate counting was performed to evaluate bacterial viability. netic field increased the percentage of cells with successful PM re- Compared with group C, bacterial viability in groups U (P=0.002), covery by 17.3 % (n=38). G(P=0.003), and T (P=0.001) were significantly decreased. Moreover, bac- CONCLUSIONS Compared with the cellular organelles that are an- terial viability was even more significantly decreased in groups U+G chored to the cytoskeleton, the freely diffusible ions of the cytoplasm (P=0.001) and U+T(P=0.001) (Fig. 2). According to Formulation 1, groups are more vulnerable to the sonoporation-induced disturbance. This U+G and U+T displayed enhanced inhibitory ratios on ESBLs E. coli bio- study directly demonstrated that the calcium influxes across the PM films of 3.7% and 2.4%, respectively. Assessment of ESBLs E. coli biofilms pores could result in cellular ionic imbalance and mitochondria mem- by CLSM. Results were analyzed using the Image J software. In all brane potential changes. This study contributes to show that static groups, the amount of living bacteria was the most in the intermediate magnetic field could protect the sonoporated cells in the scenario of layer, relatively less in the inner layer and least in the outer layer of bio- intracellular delivery of neutrally-charged drugs by reducing the films. In addition, the percentage of dead bacteria in each layer of bio- speed of ion exchanges and promoting PM resealing. films in groupsU, U+G, and U+T significantly increased when compared with the same layer of group C. This was especially true in group U+T (Fig. 3). In order to further observe the differences of biofilms morph- ology and surface viability among these groups, a 3-D surface shape of biofilms was reconstructed with pictures from CLSM. (Fig. 4) Compared with group C, thickness of biofilms in groups U, U+G, and U+T were sig- nificantly decreased (P<0.05), while group AP was significantly increased (P<0.05), indicating increased channel gaps in groups U, U+G, and U+T. Additionally, ADD was significantly decreased in groups U+G and U+T (P<0.05), suggesting the blocked nutrient supply of biofilms in these groups. With the exception of group T, TE decreased significantly in other treatment groups when compared with group C (P<0.05), indicat- Fig. 1 (abstract P59). Sonoporationinduced calcium influxes and ing the decrease of uniformity of E. coli biofilms in groups U+G, U+T, mitochondria membrane potential changes and G. Collectively, these results indicate that biofilms morphology and viability suffer damage after ultrasound treatment, either with or without antibiotics. Antibiotic penetration. To clarify the effects of ultrasound and P60 antibiotics on the permeability of ESBLs E. coli biofilms, a ciprofloxacin Effects of low-intensity and low-frequency ultrasound combined penetration test was conducted using the penetration model. This data with antibiotics on biofilms of Extended-Spectrum suggests that LILFU can increase the permeability of ESBLs E. coli Beta-Lactamases (ESBLs) producing Escherichia coli biofilms to antibiotics. Diameter of inhibition Zone (mm):C:14.0±0.3;G:21.5 1,2 Hexun Jiang ±0.4*;T:23.6±0.4*;U:26.1±0.2*;U+G:31.4±0.1*;U+T:33.6±0.3*(*Compared with Shanghai First Maternity and Infant Hospital, Tongji University School of group C, P<0.05. n=6.) Medicine, Shanghai, China; State Key Laboratory of Ultrasound CONCLUSIONS In conclusion, LILFU enhances the penetrating cap- Engineering in MedicineCo-Founded by Chongqing and the Ministry of ability of antibiotics into ESBLs E. coli biofilms, causing injuries of Science and Technology, State Key Laboratory of Ultrasound bacteria within biofilms. Ultrasound, combined with antibiotics, en- Engineering in Medicine Co-Founded by Chongqing andthe Ministry of hances bactericidal effects against biofilms. Our findings may provide Science and Technology, Chongqing, Chongqing, China a potential therapeutic method for killing bacteria insidebacterial bio- Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P60 films generated by ESBLs E. coli in vitro. OBJECTIVES The purpose of this study is to investigate the effects of low-intensity and low-frequency ultrasound (LILFU), when combined with antibiotics, on bacterial viability and the morphology of anti- biotic penetration associated with biofilms formed by extended- spectrum beta-lactamases (ESBLs) producing Escherichia coli (E. coli), a multidrug resistant bacterium strain. METHODS Experimental groups and ultrasound irradiation. The bio- films were divided into 6 groups: the control (C); gentamicin (G) with a working concentrationof 200 μg/ml; tobramycin (T) with a working con- centration of 200 μg/ml; ultrasound (U); ultrasound combined with gen- tamicin (U+G) with a working concentration of gentamicin of 200 μg/ ml; ultrasound combined with tobramycin (U+T) with a working con- Fig. 1 (abstract P60). See text for description centration of tobramycin of 200 μg/ml.The ultrasound transducer was smeared with ultrasonic coupling gel and attached to the bottom of the 6-well plate. Biofilms of groups U, U+G, and U+T were irradiatedat 42 kHz and 0.66 W/cm2 for 0.5-h, with continuous wave ultrasound. Bacterial plate count. The synergistic effect (SE) was calculated accord- ing to Formulation 1 as follow: Synergy Effect (SE) = the inhibitory ratio in the ultrasound combined with antibiotics group- the inhibitory ratio in the antibiotics group. CLSM analysis. Unattached dye was removed from the biofilms by washing. Signals were recorded using green (exci- tation wavelength at 488 nm and emissionwavelength at 515/30 nm) and red (excitation wavelength at 568 nm and emission wavelength at Fig. 2 (abstract P60). See text for description 600/50 nm) channels. Penetration of antibiotics through ESBLs E. coli Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 91 of 122 Fig. 3 (abstract P60). See text for description Fig. 1 (abstract P61). (A) Upper: The 3D cell morphology P61 reconstruction before and after ultrasound sonication. When mechanisms of intracellular gene delivery via ultrasound and DNAloaded MBs were destructed into fragments, a pore onto cell DNA-loaded microbubbles membrane was generated and the fragments were spread around 1 2 1 Ching-Hsiang Fan , Yu-Chun Lin , Chih-Kuang Yeh the site of pore; lower: another case, the debris was colocalized with Department of Biomedical Engineering and Environmental Sciences, the cell membrane after ultrasound sonication. (blue: cyan National Tsing Hua University, Hsinchu, Taiwan; Institute of Molecular fluorescence protein labeled cell membrane; red: Rhodaminelabeled Medicine, National TsingHua University, Hsinchu, Taiwan DNAloaded MBs) (B) Timelapse images confirmed the gene was Correspondence: Ching-Hsiang Fan transported to cell nuclei from DNAloaded MBs after ultrasound Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P61 sonicaton. (red: Rhodaminelabeled DNAloaded MBs; green: YFPlabeled cell membrane; blue: DAPIdye labeled cell nuclei) OBJECTIVES In the past few years, several studies on ultrasound mediated gene delivery with DNA-loaded microbubbles (MBs) is vastly expanding since it’s noninvasive and non-viral vector properties. Never- P62 theless, little is known on the machinimas of enhanced cellular gene de- Acoustic vaporisation of encapsulated droplets livery following ultrasound with DNA-loaded MBs interaction. In this 1,2 4 3 A. Podkovskiy , C. Olivier4, J.L. Thomas , M. Guedra3, F. Coulouvrat , study, we tried to image the MBs and cells during ultrasound sonication 3 1, 2 2 T. Lacour , Wladimir URBACH , N. Taulier for revealing the short-term and long-term mechanisms of intracellular 1 2 ENS Paris, Lab. Phys. Stat., Paris, France; LIB, Sorbonne Universités gene delivery, including sonoporation, endocytosis, and lipid fusion. UPMC and CNRSLIB, 75005-Paris, France; Sorbonne Universites, UPMC METHODS DNA-loaded MBs were synthesized to have cationic Univ Paris 06and CNRS UMR 7190F-75005 Paris, Institut Jean Le Rond phospholipid shell with C3F8 gas core, with DNA (green fluorescence d’Alembert, Paris, France; INSP, Sorbonne Universités UPMC 75005 Paris, protein plasmid) loaded via electrostatic interactions. The lipid shell of France DNA-loaded MBs was modified with folate acid for binding with the fol- Correspondence: A. Podkovskiy ate acid receptor of C6 glioma cell. The measured mean concentration Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P62 and size of DNA-loaded MBs were (2.1 ± 0.1) × 1010 MB/mL and 1.0 ± 0.2 μm, respectively, measured by multisizer. The DNA loading effi- OBJECTIVES The use of encapsulated droplets is currently largely in- ciency was 27.1 ± 3.1 %. Ultrasound exposure (frequency = 1-MHz, vestigated for medical applications, mainly because their reduced acoustic pressure = 0.5 MPa, cycle number = 500, pulse repetition fre- size allows them to enter targeted areas that cannot be reached by quency = 10 Hz, sonication time = 10s) was performed with presence large microbubbles (contrast agents). Perfluorocarbon droplets can of DNA-loaded MBs to achieve gene delivery with C6 glioma cells. To be vaporized under the action of an ultrasonic field, in order to turn monitor the DNA trafficking from MBs into cell nuclei, the DNA, cell them into echogeneous eventually cavitating microbubbles. The membrane and cell nuclei were fluorescently labelled with Rhodamine, optimization of acoustic droplet vaporization (ADV) will be enhanced yellow fluorescence protein (YFP) and DAPI dye, individually, and im- by understanding the physical mechanisms underlying ADV, which aged by a live-cellmulti-color fluorescent microscope before and after are currently not totally elucidated. ultrasound sonication. During the imaging, the cells were kept in a hu- METHODS A recent model (JASA p3656–3667 (2015) doi:10.1121/ midified atmosphere with 5% CO2 at 37 °C. 1.4937747) describes this phenomenon paying particular attention to RESULTS The 3-D cell morphology reconstruction data show that a pore the finite size of the droplet and its encapsulation by a thin viscoelas- (Fig. 1A, white arrows) on the cell membrane was immediately formed tic layer. Numerical simulations, done for droplets of micrometric when the DNA-loadedMBs (spherical red structure) are disrupted into radii and for frequencies of 1–5 MHz, reveal that droplet surface ten- fragments (red spot) by ultrasound sonication. We also observed that sion and shell rigidity are responsible for an increase of the acoustic the debris of DNA-loaded MBs spread around the site of pore or co-lo- droplet vaporization threshold. This threshold does not vary mono- calized with the cell membrane (white arrows), indicating that the DNA- tonically with frequency and thus an optimal frequency can be found loaded MBs vesicles might through sonoporation or lipid fusion path- to minimize it. To check the above theoretical results, we investi- ways to delivery gene into the cells. The time-lapse images (Fig. 1B) also gated in vitro the relationship between ADV and Inertial Cavitation. confirm that the gene indeed was transported and internalized into cel- RESULTS The threshold pressures required to induce ADV and IC lular nuclei from DNA-loaded MBs following ultrasound sonication. were simultaneously determined for micron-sized PFC droplets as a CONCLUSIONS This study demonstrated the intracellular DNA traf- function of droplet size and US parameters. Experimental conditions ficking from DNA-loaded MBs into cellular nuclei might rely on sono- that yield ADV without IC and ADV with IC, that could enhance drug proation, endocytosis, and lipid fusion. Future works include optimize delivery via sonoporation, are determined by investigating parame- the parameters of ultrasound and DNA-loaded MBs for improving ters that could influence both thresholds; bulk fluid properties such the efficiency of gene delivery. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 92 of 122 as gas saturation, temperature, viscosity, and surface tension; droplet P64 parameters such assurfactant type; and acoustic properties such as DOX-loaded superhydrophobic mesoporous silica nanoparticles pulse repetition frequency and pulse width. with ultrasound for dual-treatment in tumour CONCLUSIONS In conclusion, our experiments allowed us to test a QiaoFeng Jin, Cheng-Han Wu, Chih-Kuang Yeh new model of droplets vaporization using ultrasound in an infinite Department of Biomedical Engineering and Environmental Sciences, medium. The study of droplets vaporization, located in confined National Tsing Hua University, Hsinchu, Taiwan medium more resembling the situation in vivo, will be the next step. Correspondence: QiaoFeng Jin Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P64 OBJECTIVES Mesoporous silica nanoparticles (MSNs) have great P63 potential for biomedical applications. Surface air bubbles Ultrasound theranostics for tumor adsorbed on the fluorine-modified superhydrophobic MSNs could Kazuo Maruyama be served as cavitation nuclei and prevent the leakage of chemo- Faculty of Pharma-Sciences, Teikyo University, Tokyo, Japan therapy drugs encapsulated into MSNs during circulation. Com- Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P63 bination of superhydrophobic surfaces with ultrasound can cause cavitation to induce cell damage and facilitate drugs release. In OBJECTIVES “Theranostics” is a treatment strategy that combines this study, we propose a dual-treatment strategy by combining therapeutics with diagnostics. The combination of bubble formula- active cavitation and chemotherapy. tion and ultrasound (US) is a good tool for “theranostics” due to have METHODS Synthesis of superhydrophobic F48-ext and Dox loaded multi-potency both of diagnostics with enhanced echo signal from F48-extThe parent MCM-48 type MSNs were synthesized with sol- bubble and therapeutics with cavitation of bubble. To develop a gel method using BCDAC and C16E2 as soft templates. To fabri- novel bubble formulation for US imaging and therapy with small par- cate fluorinated MCM-48 (names as F48-ext), the MCM-48 were ticle size and a good stability and test the formulation as US imaging dispersed in 10 mL toluene containing 0.2 ml of perfluorodecyl- contrast agent and for gene delivery in vitro and in vivo. Interleukin12 triethoxysilane and stirred at 100° for 48 hours, and the resulted (IL-12) exhibits immunomodulatory antitumor effects and is consid- solid was collected and dried at 60° for 12 hours. Finally, F48-ext ered an effective antitumor agent, but its short half-life and systemic NPs were obtained after removing the surfactants by repeated toxicity following intravenous injection are major obstacles to its ion exchange in a dilute HCl-ethanol solution at 60°. The DOX therapeutic use. Therefore, we transfected pDNA encoding the IL-12 was loaded into F48-ext NPs in ethanol which contains 2% (v/v) gene (pCMV-IL-12) into tumor tissue using bubbles and US with the concentrated hydrochloric acid, then the ethanol were vaporized aim of achieving high local expression of IL-12. and replenished with 1ml ethanol contain 2% hydrochloric for METHODS Lipid-stabilized bubbles were prepared by homogenization thrice. Inertial cavitation dose measurement using a passive cavi- of a lipid dispersion in the presence of perfluoropropane gas. Different tation detectorA 15-MHz focused transducer was used as a pas- phospholipid compositions were tested and evaluated. After bubble sive cavitation detector to receive inertial cavitation (IC) signals, formation the bubbles were freeze-dried so that a dry sample contain- while a 2-MHz HIFU transducer was used to excite theNPs. The ing bubbles was formed. After re-constitution of the samples they were integrated value of the amplitude of frequency spectrum from analyzed for size, gas content and US signal intensity. The ultrasound 10-20 MHz was termed inertial cavitation dose (ICD) and used to theranostics capabilities of bubbles for the solid tumor were studied in assess the IC intensity. In vitro cell cytotoxicity with alamar Blue® Colon26 tumor bearing mice. Bubbles was injected to mice via tail vein with and without US exposureTRAMP cells were incubated with and 9 MHzlinear ultrasound was exposed to solid tumor site transder- Dox-loaded F48-ext NPs, free F48-ext NPs and free Dox for 2 mally. Following the recognition of neovasculature in tumor tissue, 1 hours in 24 wells plate, respectively and then all samples were MHz therapeutic ultrasound was exposed transdermally over the site of sonicated at 5 MPa pressure with a PRF of 100 for 5 min by the solid tumor tissue. 2-MHz transducer. The cells without ultrasound sonication were RESULTS Bubbles was injected to mice via tail vein and 9 MHz linear set as control group. After 3 hours, the samples were washed ultrasound was exposed to solid tumor site transdermally. The flow and further incubated for 24 hours, and their cytotoxicity were of bubbles in blood was observed and neovasculature of tumor tis- measured by using alamar Blue® kit. sue was imaged clearly. Following the recognition of neovasculature RESULTS MCM-48 and F48-ext NPs show average sizes of 200-300 in tumor tissue, 1 MHz therapeutic ultrasound was exposed transder- nm, and show the zeta potentials of -30 mV, which contributed to mally over the site of solid tumor tissue. This process induced cavita- the stability of such NPs. The F48-ext NPs had a contact angle of tion of bubbles in the tumor tissue, resulted in rising the about 150°, and after loading Dox their contact angle decreased to temperature of tumor tissue to 45-55C, and also significant reduction 130° due to the hydrophilic Dox. The ICD of F48-ext NPs have signifi- of tumor growth. Cavitation leads to localized heating and cloud be cantly augmented with respect to that of MCM-48 NPs (Fig. 1). The use for ablative cancer therapy. Transfection of pCMV-IL-12 with LBs ICDs of F48-ext and Dox-loaded F48-ext NPs both increased with and US suppressed tumor growth significantly. To investigate the acoustic pressures. Due to the superhydrophobic modification, the mechanism behind the anti-tumor effects of pCMV-IL-12 transfected Dox leakage of F48-ext NPs was suppressed. We found that Dox re- using bubbles and US, we assessed the involvement of CD4+ and leasing of F48-ext NPs was not increased with ICD increasing. For the CD8+ T cells and NK cells. The depletion of CD8+ T cells effectively short term, the cells damage caused by inertial cavitation of F48-ext blocked the antitumor effect of pCMV-IL-12 transfected using bub- NPs was observed and dominated the cell cytotoxicity. The Dox re- bles and US. These results suggest that the combination of bubbles leasing from the F48-extNPs performed long term treatment mode. and US can effectively induce sufficient IL-12 expression to cause CONCLUSIONS In summary, the degree of inertial cavitation of MSNs anti-tumor immune responses. increased after fluorine-modification (F48-ext), and the Dox-loaded CONCLUSIONS The combination of bubbles and US could be effica- into F48-ext NPs shows better stability than MCM-48 NPs. In addition, cious for neovasculature image and cancer therapy. We believe this DOX loading did not inhibit the IC activity of F48-ext NPs. Combin- new formulation shows great promise for both diagnostic and thera- ation of mechanical effect and chemotherapy have shown the tumor peutic applications thanks to its good stability, relatively small bub- dual-treatment mode. We demonstrated that F48-ext NPs were sensi- ble size and the simplicity of handling. tive to ultrasound and with a stable drug-loading capacity. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 93 of 122 OBJECTIVES To give advice about how to select suitable patients in MR-guided focused ultrasound surgery (MRgFUS) of uterine fibroid by screening pelvic MRI. METHODS Retrospective analysis the MRI image and clinical data of 30 fibroid patients who successfully complete the MRgFUS treatment in Huashan hospital and also review the literature. RESULTS The safe and effective use of MRgFUS is affected by fibroid type and location, position relative to adjacent anatomical structures and the presence of coexistent pelvic disease. CONCLUSIONS Screening pelvic MRI for selection of patients in whom sufficient fibroid volumes can be treated safely using the MRgFUS system is critical forsuccessful outcomes. Fig. 1 (abstract P64). (A) The diagram of controlled drug release P67 from Dox laoded superhydrophobic F48-ext with air layer; (B) the Effect of fiducial markers and brachytherapy seeds on the delivery experimental setup of the ICD measurement; (C) the contact angle of HIFU salvage therapy in the prostate 1 2, 1 3, 1 3 of the MCM-48, F48-ext with and without Dox; (D) the ICD of the Panayiotis S. Georgiou , Jiri Jaros , Heather Payne , Clare Allen , 4 4 1 1 MCM-48, F48-ext and F48-ext Dox; (E) the compare of cell cytotoxicity Taimur T. Shah , Hashim Ahmed , Eli Gibson , Dean Barratt , of F48-ext dox with and without ultrasound exposure Bradley Treeby Medical Physics and Biomedical Engineering, University College London, London, UK; Computer Systems, Brno University of Technology, Brno, CzechRepublic; Oncology, University College London Hospitals, London, UK; Surgery and Interventional Science, University College London, London, UK P65 Correspondence: Panayiotis S. Georgiou The influence of exercise on uterine fibroid and adenomyosis after Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P67 high-intensity focused ultrasound Huang Xueyan OBJECTIVES Prostate cancer is the most common cancer and the sec- Biomedical Engineering, Chongqing, China ond leading cause of cancer-related death in men in Europe and North Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P65 America. For patients with early-stage localized disease, the cancer is often treated using external beam radiation therapy (EBRT) or brachyther- OBJECTIVES To evaluate exercise can improve clinical symptoms of apy. In EBRT the radiation treatment is delivered from a source external uterine fibroid and relief dysmenorrhea of adenomyosis after high-in- to the patient’s body and the procedure usually involves implanting a tensity focused ultrasound (HIFU) treatment. small number of gold fiducial markers into the prostate to verify the pos- METHODS From January 2011 to August 2015. 83 patients ition of the prostate gland between treatments. In brachytherapy the ra- suffered from uterine fibroid and 102 patients suffered from diation dose required for the treatment is delivered locally by implanting adenomyosis. All patients were treated withHIFU.Among the 185 a large number of radioactive seeds (approximately 100). In low dose rate patients, 83 of them completed one-year follow-up,102 patients (LDR) brachytherapy these seeds remain permanently in the patient’s of them completed half a year follow up. All the symptom im- body. For some of these patients, treated either by EBRT or LDR brachy- provement the uterine fibroid volume and the standards of Visual therapy, their cancer will recur. In such cases, further treatment using an Analogue Scale(VAS) were recorded. alternative (salvage) therapy can be considered. High intensity focused RESULTS In this study, 83 patients had uterine fibroid,51 of 83 ultrasound (HIFU) is currently offered in hospitals as a minimally invasive patients were no exercise intervention group, the rest of 83 pa- salvage therapy for treating prostate cancer inpatients whose cancer has tients were exercise intervention group,Comparing with this two recurred. However, clinicians observe mixed results with salvage-HIFU for groups’ uterine fibroid change,P=0.003,considered as statistically failed brachytherapy, which may be partly linked to inadequate heating significant.That meaning, exercise could help improve the absorp- caused by the implanted seeds. To date, the impact of implanted EBRT tion of uterine fibroid.102 patients had adenomyosis,45 of 102 markers and brachytherapy seeds on HIFU treatment has not been thor- patients were no exercise intervention group,57 of 102 patients oughly studied. The objective of this work was to investigate the effect of were exercise intervention group.From the result found that exer- a single EBRT marker on the efficacy of HIFU treatment delivery and ex- cise may relief the score of dysmenorrhea,P=0.002,considered as tend these results to a large number of brachytherapy seeds. statistically significant. METHODS Using the k-Wave acoustics toolbox, we have performed CONCLUSIONS Based on the observations from uterine fibroid coupled acoustic-thermal modelling of HIFU treatments by first model- and adenomyosis, exercise have positive reaction on improving ling the prostate with the presence of a single EBRT gold marker at dif- symptoms of uterine fibroid andadenomyosis. Hence, this clinical ferent positions and then, with the presence of multiple brachytherapy effective and response is a further treatment for woman who seeds. The medium properties were obtained from book values. The wants to protect the uterus. transducer model was based on the specifications of the Sonablate 500 HIFU probe. The distribution of the seeds was extracted from patients’ medical images. The target lesion volume, location and order of sonica- P66 tions were selected based on standard treatment protocols. The ab- Patient selection by screening pelvic MRI in MR-guided focused lated lesion volume was estimated from the model data using a ultrasound surgery of uterine fibroid thermal dose metric based on cumulative equivalent minutes. 1 2 1 1 1 Junhai Zhang , Xiaoxia Liu , Hairui Xiong , Zhenwei Yao , Qian Zhou , RESULTS The simulation results show that the EBRT marker obstructs 1 1 Haoxiong Li , Ying Tang the propagation of the ultrasound beam and distorts the focal Department of Radiology, Huashan Hospital, Fudan University, volume when positioned within 5mm of the focus (Fig. 1). The distor- Shanghai, China; Department of Gynecology, Obstetrics & Gynecology tion significantly increases when a large number of seeds is intro- Hospital, Fudan University, Shanghai, China duced in the model. Both the seeds and the markers act as Correspondence: Junhai Zhang scatterers, reflecting energy away from the intended focus. This leads Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P66 to two undesired implications. Firstly, the intended treatment region Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 94 of 122 receives less energy and as a result thetissue is not adequately thrombosis; three patients reported lower extremities numbness, two heated. Secondly, the wave reflected by the seeds induces heat- patients had soft tissue edema around the lesions. ing at unintended regions, which often include healthy tissue CONCLUSIONS MRgFUS is effective for short-term pain palliation of and the rectal wall. bone metastases. Such noninvasive technique is safe and can im- CONCLUSIONS The distortion introduced due to the presence of the prove patients’ living condition. EBRT marker, and more importantly, due to the presence of the brachy- therapy seeds may result in undertreated regions due to less energy ar- riving at the focus or overtreated regions due to reflections which may affect organs at risk. Depending on the position of the targeted regions P69 and the distribution of the markers or seeds, both effects may be un- Characterization and validation of a 64-element Capacitive Micro- desirable. The results presented have particular importance for patient Machined Ultrasound Transducer (CMUT) annular array with a 256 selection and treatment planning for prostate salvage-HIFU after failed element imaging array in the same plane, capable of tissue EBRT or brachytherapy and indicate the necessity to reconsider the ablation of the prostate 1, 2 1, 2 1 treatment parameters used during theseprocedures. Christopher Bawiec , William Apoutou N'Djin , Guillaume Bouchoux , 3 4 1, 2 Nicolas Sénégond , Nicolas Guillen , Jean-Yves Chapelon 1 2 Inserm, U1032, LabTau, Lyon, France; Université Lyon 1, Lyon, Rhone- 3 4 Alpes, France; Vermon, Tours, France; Edap TMS, Vaulx-en-Velin, France Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P69 OBJECTIVES The goal of this work was the validation and characterization of a recently developed CMUT probe that is capable of delivering a therapeutic ultrasounddose while simultaneously en- abling the delivery to be monitored through ultrasound imaging. The primary application of this probe is the treatment of prostate cancer forthermally ablating the cancerous tissues. The currently used probe Fig. 1 (abstract P67). See text for description is a high intensity focused ultrasound (HIFU) transducer utilized treat- ment of prostate cancer. Thesedevices are fabricated using spheric- ally focused piezoelectric materials, however, there are drawbacks to the current design. In this work, CMUT probes were investigatedfor P68 the HIFU applications as they offer potential advantages over the The safety and short-term efficacy of MR guided focused current designs allowing decreased fabrication costs at an industrial ultrasound surgery for bone metastases-induced pain palliation scale and ease ofminiaturization. 1 1 1 1 1 Hairui Xiong , Qian Zhou , Junhai Zhang , Haoxiong Li , Ye Chen , METHODS A planar ultrasound probe consisting of a 64-element CMUT 2 1 1 1 Qiong Li , Ying Tang , Zhenwei Yao , Xiaoyuan Feng annular array around a linear 256-element CMUT imaging array was Department of Radiology, Huashan Hospital, Fudan University, studied in simulation and experimentally (impedancemetry, microscopy Shanghai, China; Department of Anesthesiology, Huashan Hospital, and vibrometry as well as hydrophone measurements and in vitro le- Fudan University, Shanghai, China sion formation). Numerical models of the therapeutic acoustic field Correspondence: Hairui Xiong were performed using the Rayleigh integral to determine the feasibility Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P68 of electronic focusing and inducing thermal lesions. For comparison, simulations of a geometrically focused existing piezo-based technology OBJECTIVES To discuss the safety and short-term efficacy of MR- (16-element annular array) was also performed. Experimentally, key guided Focused Ultrasound Surgery (MRgFUS) for pain palliation of static parameters (collapse-, snapback-, andbreakdown voltages) for bone metastasis patients. use in the different modes of operation (conventional, collapse, and METHODS 14 patients with painful bone metastases were recruited collapse-snapback) were identified. The HIFU capabilities of the device in a prospective study. The treating efficacy is characterized by Nu- were also investigated experimentally (pressure field and radiation merical Rating Scale (NRS), theBrief Pain Inventory Quality of Life force measurements) with the creation of in vitro lesions. The imaging (BPI-QOL) survey, and Karnosky Performance Status Scale (KPS). the capabilities were also compared between the existing device and the adverse events occurred pre and post-treatment were analyzed. Nor- prototype in order to validate the modelling. mal distributed statistics were analyzed with paired-samples t test, or RESULTS Simulations showed that the planar CMUT design could Wilcoxon rank sum test was used. dynamically focus from 3-7cm and create lesions comparable in RESULTS 14 patients were treated with MRgFUS. two patient size and shape to the geometrically focused transducer. Microscopy dropped out of the study. The NRS ratings are 6.5(4), 5(5.25), 2.5(5), allowed visualization of the static collapse and snapback of individ- 2.5(4.75) for pre-treatment, one week, one month, two months, and ual cells which was confirmed in parallel with impedance measure- three months, respectively. Such variances of NRS ratings are statisti- ments. Collapse occurred at 132.5±5V and snapback occurred at 95 cally significant (P<0.05). The BPI-QOL ratings are 42.42±8.27,30.67 ±5V. Dynamic behavior of the CMUT (vibrometry) in air and in ±12.29, 29.17±15.38, 29.92±17.67, 35.67±19.28, respectively. The BPI- water allowed identification of the collapse-snapback operational QOL ratings decrease in the first two months after the treatment mode. In this mode, the probe was capable of generating up to which is statistically significant (P<0.05) ; whereas for the third 5W/cm2 surface intensity. Furthermore, experimental validation month, the BPI-QOL rating is statistically insignificant compared with performed on the probe utilizing hydrophones and in vitro tissue the one before the treatment. The KPS ratings are80(28), 80(20), confirmed that the probe was capable of creating lesions in tissue. 65(45) for pre-treatment, one week and three months, respectively. CONCLUSIONS The results of the investigation confirmed that the Three months after the treatment, the KPS ratings decreases which is current prototype is capable of creating lesions in biological statistically significant compared with the one before the treatment tissue. Both the simulations and the experiments were able to (P<0.05). After the treatment, one patient developed deep venous confirm the capability of this ultrasound probe. The ultrasound Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 95 of 122 imaging part of the probe also provided an enhancement to the experiment conditions. The system with the software light field can spatial resolution of the existing technology which allowed for a measure the average intensity of sonoluminescence. higher resolution improving the localization of the treatments. RESULTS The intensity of multibubble sonoluminescence was in- This new ultrasound probe prototype has the potential of provid- creased, when ambient pressure increased from 0.1MPa to 10MPa. ing an improved treatment method of prostate cancer that can However, the region of cavitation cluster decreased gradually. increase the quality of the treatment and subsequent outcome of CONCLUSIONS The increasing ambient pressure caused active the patient while adding the benefits of a reduction in the cost area of cavitation cluster to decrease. Besides the cavitation clus- and size of the probe. This project was supported by the French ter activity will be more drastic, when electronic power reaches a Single Interministerial Fund (FUI, 2013). certain value. P70 P72 Precision microvascular therapy via the synergy of light and sound 1,2 2,3 2,3 3 3 4 5 Method of temperature estimation in skin for a focused ultrasound H. Zhang ,Z. Hu ,J.Li , Q. Liu , S. Yuan , Y. Paulus , X. Yang , 1,2 applicator using a multi-validated numerical thermal model and X. Wang Aghuinyue E. Umenei, Ziqi Wu Institute of Acoustics, Tongji University, Shanghai, P.R. China; Global Discovery, Amway, Grand Rapids, Michigan, United States Department of Biomedical Engineering, University of Michigan, Ann Correspondence: Aghuinyue E. Umenei Arbor, MI, USA; Department of Ophthalmology, the First Affiliated Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P72 Hospital of Nanjing Medical University, Nanjing, P.R. China; Department of Ophthalmology and Visual Sciences, University of Michigan, Ann OBJECTIVES Focused ultrasound through ablation has gained increasing Arbor, MI, USA; Department of Mechanical Engineering, University of use in aesthetic applications over the last decade. Recent studies show Kansas, Lawrence, KS, USA non-ablative focused ultrasound also provided clinical benefits through Correspondence: H. Zhang localized heating. Temperature measurements are critical for efficacy and Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P70 safety for such applications but are not easily or cheaply obtainable under in vivo dermatological conditions. Current temperature measure- Angiogenesis and neovascularization are hallmarks for a variety ment methods are expensive, time consuming, application specific/organ of pathological conditions, including cancer and many eye dis- dependent and provide unsatisfactory spatiotemporal resolution. We eases, and play a crucial role in disease onset and progression. propose a method including a multiphysics model to quickly estimate Antivascular therapies that aim at either removing microvessels skin temperature distribution in 2D, which was validated using multiple or slowing down their growth represent a proven new strategy approaches and lends itself to an iterative development process. to intervene the progress of these conditions and improve the METHODS An acousto-thermal 2D FEA multiphysics model was used prognosis. Here we report the development of a photo-mediated to estimate focused ultrasound thermal distribution in a tissue mimic ultrasound therapy (PUT) technique as a new concept of localized materials (TMMs) which are acoustically similar to skin tissue, by in- antivascular therapy. Unlike any of the previous optical- or corporating acoustic field 2D surface measurements of the 3W appli- ultrasonic-alone techniques, laser pulses and ultrasound bursts cator from its hydrophone scans (Fig. 1). The effects of ultrasound on are synergistically applied in PUT, which facilitate noninvasive tissue heating were modeled in two phases with parameters both treatment of subsurface microvessels with unprecedented high measured in the lab and obtained from literature where possible. Ini- precision. PUT takes advantages from the high native optical con- tially, the acoustic pressure in theTMM generated by the applicator trast among biological tissues, and has the unique capability to was initially solved using wave equations in the frequency domain. self-target on blood vessels without causing unwanted damage The thermal distribution in the tissue was subsequently calculated to surrounding tissue. As demonstrated through the experiments using the obtained acoustic energy as the heat source term to the on animal models, PUT can treat microvessels in target tissue via bioheat equation in a time domain simulation over 7 seconds (Fig. 2). different mechanisms, including blocking local vessels by indu- The simulated transversal 2D temperature distribution was compared cing blood clots or disrupting vessels causing local hemorrhage, to thermocouple measurements in the tofu for multiple applicators each with values in clinic. Moreover, PUT working at different (10). The thermocouple measurements were taken at 0.5mm spacing optical wavelengths can selectively treat veins or arteries by along the longitudinal axis of the applicator, at 2, 4, 6, and 8 mm below utilizing the contrast in optical spectra between deoxy- and oxy- the surface to give a 2D measurement grid of the TMM. The focal dis- hemoglobin. PUT, as a novel antivascular therapy technique with tance of the ultrasound applicators were measured to be 4±1 mm in the capability to precisely target vessels and precisely control the water. Furthermore, simulations were performed using a porcine treatment effects, holds promise to impact clinical management muscle model and the results were compared in a similar fashion to of cancer and eye diseases by delivering optimized treatment MR thermometry acquired in muscle samples. The validated multiphy- outcome with minimized complication. sics model was finally used to estimate a thermal distribution for multi- layered skin model. RESULTS The model estimated average maximum temperature rise from nominal (37 C) in tofu at the focus, at 6.9 C compared to 6.56 C from the P71 thermocouple measurement along the focal line of interest. A good cor- Effect of ambient pressure on cavitation bubbles at the focal point relation (R2 = 0.89) was seen between modeling results and thermo- of a spherical resonator with open ends couple measurements in tofu for 10 ultrasound transducers (Fig. 3). For 1,2 1, 2 1 1, 2 Zonggui Chen , Huan Liu , Xiaobo Gong , Faqi Li porcine muscle, model estimated temperature rise along focal line of State Key Laboratory of Ultrasound Engineering in Medicine Co- interest of 14.9 C whereas MR thermometry measured 14 C. Model accur- Founded by Chongqing and the Ministry of Science and Technology, acy was between 88%-93% once equipment error was factored, provid- Chongqing, China; College ofBiomedical Engineering Chongqing ing an acceptable temperature estimate for developmental research at Medical University, Chongqing, China the high resolutions need for skin tissue. Skin multilayer model estimated Correspondence: Zonggui Chen the maximum temperature was 19 C at the focal region. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P71 CONCLUSIONS Model estimated temperature in tissue mimic mater- ial and porcine muscle were validated with thermocouple and MR OBJECTIVES Study on the activity of cavitation cluster under the differ- thermometry approaches. Byinputting free-field acoustic field infor- ent ambient pressure by the average intensity of sonoluminescence. mation and acousto-thermal properties of tissue media and building METHODS The experiments, activities of cavitation bubbles and the a validated 2D model, we have the proposed an easy method of esti- sonoluminescence are captured by general camera and emICCD mating temperature changes in multilayered tissue such as skin with under different ambient pressure, are conducted in the same close approximation at a low cost and good accuracy. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 96 of 122 P73 A 3D-printed skull model with corresponding acoustic characteristic of human skull for ultrasound brain imaging and diagnosis Chen Bai, Meiling Ji, Dui Qin, Mingxi Wan The Key Laboratory of Biomedical Information Engineering of Ministry of Education, Department of Biomedical Engineering, School of Life Science and Technology, Xi`an Jiaotong University, Xi'an, China Correspondence: Chen Bai Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P73 OBJECTIVES Recently, more and more attention has been paid to the ultrasound brain imaging as it is non-radioactive compared with CT and low-cost compared with MRI. And the existence of skull limits the propagation of the ultrasound and then limits the frequency of transducer to be transmitted and furtherly limits the resolution ofthe resulting image. To realize the noninvasive transcranial brain im- aging, the foreknowledge of the acoustic characteristic of the skull is essential. The ossa temporale (temporal bone), which is the thinnest part, has been proved that is the best acoustic window for imaging. However, the intact human skull was precious and hard to get the part we just need by excision. Instead, with the help of the 3D print- ing technology and related software, this problem can be overcome easily. Furthermore, various invitro research experiments about brain Fig. 1 (abstract P72). MRI image of applicator in pork muscle (axial imaging can be proceeded with the 3D-printed model which section) after 7s showing thermal profile, and calculated temperature rise matches the acoustic characteristic of human skull. METHODS The acoustic attenuation coefficient and ultrasonic sound velocity for the temporal bone was firstly measured. The human skull was mounted on a threedimensional scaffold and immersed in a tank with degassed water. Two single element transducers were placed on the walls of the tank coaxially on each side of the skull manually parallel to the temporal bone, one for transmitting signal and the other for receiving. The transducer was excited by a 5-cycle Sine sig- nal whose peak to peakvalue was 2 V and the receiving signal was collected with the NI data acquisition system. The direct ultrasonic wave in free field was normalized and recorded as a reference signal for calculating the attenuation coefficients and velocity. Each time, we measured at 10 different points of the temporal bone, and each point was repeated ten times and get the mean value (Fig. 1). For ensuring the optimum frequency for transcranial experiment, 5 pairs of transducers with frequency ranging from 1 to 5 MHz were tested. Moreover, the CT scan images of human skull was obtained, with 1mm interval between 2 layers, to measure the thickness of bone, and furtherly, to design and modify the model by MIMICS for 3D print. After that, to find a certain material matching the acoustic characteristics of the human skull and the cerebral vessels, the at- tenuation coefficients and velocities of 9 different kinds of materials including resin, nylon, PEEK, PLA, etc. with different thickness were measured. Additionally, considering the effect of the temporal bone’s curvature, the materials were modeled in the shape of the temporal Fig. 2 (abstract P72). Simulated thermal model of TMM (tofu), bone rather than flat plate. By aid of the software MIMICS, we got the showing ultrasound applicator and temperature distribution after satisfying models with excision of the redundant part of the skull. 7s (longitudinal section) RESULTS According to the results of measurement, 1.8 to 2 MHz was determined with synthetic consideration of the attenuation, propagation and the resolution for further imaging. Hence, the averaged attenuation coefficient and velocity of the temporal bone for the human skull model measured with 1.9 MHz frequency were 7.34 dB/mm and 2398.13 m/s, respectively. And for the optimum material within the 9 kinds of mate- rials, PLA, those were 7.11 dB/mm and 2270.24 m/s under the same test- ing condition, which means the thickness of the 3D-printed skull model with the PLA should be the same as that of human skull. In other words, the mean thickness of the temporal bone of the printed skull model using PLA was about 1.8 mm, in correspondence with the acoustic char- acteristic of human skull with the mean thickness of1.7 mm. For another, the material to print the vascular model, one kind of resin, the SPR6000B, with averaged attenuation coefficient 0.8 dB/mm and velocity 1963 m/s, approximating to the results in the free field, was picked out. Finally, by Fig. 3 (abstract P72). Surface map and line measurement of measuring the performance of the new models, it has been proved that temperature rise of ultrasound applicator in tofu after 7s the human skull and cerebral vessels can be conveniently and com- (longitudinal section) pletely replaced by the 3D-printed models. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 97 of 122 CONCLUSIONS In order to design a skull model which can completely transducer with periodic array surface was designed to realize replace the intact human skull in ultrasound brain imaging study, we focusing enhancement. measured the acoustic characteristics of temporal bone of human skull, METHODS To verify the enhanced acoustic focusing of concave and moreover, a number of different materials to ascertain the material transducer with periodically aligned subwavelength grooves, the whose acoustic attenuation characteristic was similar to the human acoustic pressure and temperature elevation generated by the new skull or the vascular. The results showed that PLA met the requirement structure were investigated based on both experimental measure- for skull model and the attenuation coefficient of SPR6000B was small ments and numerical simulations, and then compared with the re- enough to print the vascular model. Finally, we have acquired the satis- sults obtained from conventional concave transducer. The software fying skull model with excision of the cranial skullcap, and to fix the de- of Comsol Multiphysics was employed to perform the numerical sim- signed vascular model conveniently. In the following study of brain ulations based on the combination of2D Helmholtz equation and imaging, a scheme that has the contrast agents circulate in the vascular bio-heat transfer equation. An optical fiber hydrophone and thermo- model fixed on the skull model, and manage to imagethe vascular couple were utilized to measure the acoustic field and detect the through the temporal bone via a 128-element linear array transducer temperature rise at the focus in a tissue phantom exposed to HIFU with 2 MHz frequency will be performed. pulses, respectively. RESULTS Both the experimental measurements and numerical simu- lations show that the peak pressure amplitude and the rate of temperature rise at the focal region are obviously larger in the case of corrugated transducer rather than the conventional transducer as Fig. 1. Different from previous studies originated from planar arrays of periodic structure, the application of spherically curved arrays of periodic grooves could result in extraordinary acoustic transmission close to Wood’s anomaly that at the wavelength slightly less than the groove periodicity. CONCLUSIONS Numerical simulations and experimental measure- ments both show that, comparing with conventional concave trans- ducer, the concave transducer with periodic array of subwavelength grooves is more efficient to improve acoustic focusing and enhance relative bioeffects. This work may open new possibility to design more favorable focused ultrasonic transducers that could significantly improve HIFU treatment effects. Fig. 1 (abstract P73). Receiving signals in the (a) free field, (b) skull, (c) PLA and (d) SPR6000B; (e) the CT scan image and (f) the 3Dprinted model Fig. 1 (abstract P74). Simulated acoustic pressure lever and temperature distributions on the focal plane: acoustic pressure lever with (a) the corrugated transducer and P74 (b)conventional transducer; and temperature distribution in Enhanced ultrasonic focusing and temperature rise by using phantom exposed with (c) the corrugated transducer and (d) sub-wavelength periodic structure conventional transducer Chenghai Li, Xiasheng Guo, Juan Tu, Dong Zhang, Zhou Lin Key Laboratory of Modern Acoustics (Nanjing University), Ministry of Education, Nanjing University, Nanjing, Jiangsu, China Correspondence: Chenchen Bing Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P74 P75 Optimization of thermal fields by focal region modulation in HIFU OBJECTIVES High-intensity focused ultrasound (HIFU) has been brain tumour treatment used in clinic as a non-invasive therapeutic method to treat solid 1 2 1 1 Shihui Chang , RUI CAO , Yabin Zhang , Shijing Wu1, Xiqi Jian tumors for decades. Although various (e.g., lens, spherically 1 2 Tianjin Medical Univercity, Tianjin, China; Tianjin University of Science & curved transducer and multi-element phased array) were devel- Technology, Tianjin, China oped to focus ultrasonic energy, the ultrasonic focusing efficiency Correspondence: Shihui Chang is highly restricted by the size of transducers due to the limita- Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P75 tion of manufacture technology. In the present work, a concave Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 98 of 122 OBJECTIVES The temperature rise during HIFU brain tumor treatment convective cooling and acoustic streaming can change the should be controlled accurately to avoid side effects by overheating temperature considerably near blood vessel. such as cerebral hemorrhage. In this work a modulation method to CONCLUSIONS Three dimensional simulations of focused ultrasound control the shape and volume of focal region as well as the ablation has been perfomed in a patient specific geometry. Simula- temperature distribution at focus was performed by adjusting the tion using multiple GPUs can sufficiently reduce the simulation time driving signals of the transducer elements. and can help to construct surgical planning platform. High ultra- METHODS The simulation model was reconstructed based on the CT sound power and nonlinear propagation effects with appropriate data of a volunteer's head and the 82-element transducer. In this treatment planning can sufficiently reduce the treatment time. We study, the finite difference time domain (FDTD) method was used to theoretically showed that the treatment time can be reduced to few calculate the temperature distribution induced by HIFU in the brain. minutes. The presented model can be used in planning tools for the Two foci with a certain distance were generated on the acoustic axis thermal ablation of tumour in other organs. by a HIFU source using two driving signals which were determined by time reversal method. This double foci fusion method was applied to create a uniform temperature distribution at focal region and ad- P77 just the length and volume of the focal region. Principal component analysis of acoustic aberrations for rapid RESULTS Thermal field distribution at focal region can be ad- HIFU beam refocusing: a simulation study justed by changing the time delay of the two driving signals. Xiaoxu Lei, Martijn de Greef, Chrit Moonen, Mario Ries Thus, a uniform distributed temperature field can be created at Imaging, University Medical Center Utrecht, Utrecht, Netherlands focal region. What is more, this method was able to adjust the Correspondence: Xiaoxu Lei length andvolumeofthe focalregionusing thesametotal Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P77 acoustic energy by varying the distance between the two foci. CONCLUSIONS The HIFU temperature distribution, shape and volume OBJECTIVES The propagation path of the HIFU beam frequently in- of the focal region could be modulated by the double foci fusion troduces acoustic aberrations, which degrade the focus quality. Non- method. The focal region volume formed by a single irradiation could invasive refocusing 1-3 based on acoustic radiation force imaging be enlarged effectively while avoiding the brain tissue overheated. This (ARFI) have been proposed, which address this problem. However, method can provide basis for the focal temperature distribution modu- the refocusing process is generally lengthy and cumbersome, due to lation in the treatment of brain tumor in further study. the multitude of independent measurements. Here, we evaluate the feasibility of accelerating this process by replacing the Hadamard (H) base and Zernike polynomials (ZP) base of the original approaches P76 by a new base which is derived using principal component analysis Modelling and simulation for the focused ultrasound ablation of (PCA) on simulated phase aberrations from a uterine fibroid model. liver tumour METHODS The pressure at HIFU focus can be expressed as the sum 1,2 2 1 1 Maxim Solovchuk , Tony W.H Sheu , Manuel Diaz , Peter Deng of the product of the acoustic wave fired from each individual trans- Institutes of Biomedical Engineering and Nanomedicine, National ducer element and the acoustic aberration experienced along its 2 N Health Research Institutes; Engineering Science and Ocean Engineering beam path:p = ∑ =1 gi.χi, (1)where χi is the complex source term of Department, National Taiwan University each emitter, gi the aggregated attenuation and phase shift along Correspondence: Maxim Solovchuk the ith propagation path for the element and p the complex focus Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P76 pressure of all elements.For non-iterative adaptive refocusing approach, Eq.1 can be transformed into matrix form:pX = g.X, OBJECTIVES High intensity focused ultrasound is very promising new (2)Where pX is a 1 x K complex row vector whose elements are the technology, that has many therapeutic application, among them are focus pressures corresponding to each excitation pattern from base the treatment of cancer indifferent organs. One of the main diffi- X, g is a 1 x N complex row vector containing the aberration cultlies, that limits further development of HIFU, is very difficult treat- information for each transducer element, X is the N x K orthogonal ment planning and unpredictable behavior of the necrosed area in base whose column vectors are used for the actual measurement the presence of bubbles. Computational fluid dynamics can greatly process, which requires 4xK independent measurements. Both, a help in this regard. Calculation of ultrasound propagation in a patient Hadamard base H and Zernike polynomials base Z have been specific geometry is very time consuming process. In order to applied as X in Eq.2 for refocusing the beam. In order to accelerate achieve reasonable simulation time HPC (high performance comput- the refocusing process, the less numbers of columns of X used to get ing) should be used. The problem will be solved using HPC on mul- accurate g in Eq.2 the better. Here, a method is proposed to apply tiple GPUs. This work is a step towards the development of surgical PCA on prior knowledge of the phase aberrations obtained from planning platform for a non-invasive HIFU tumour ablative therapy in acoustic simulations to form an orthogonal base (PCA base), which a real liver geometry. allows to describe the dominating phase aberrations in a more METHODS The computational model has been developed for the compact form and thus allows to significantly reduce the prediction of temperature elevation in the patient specific liver measurement process. Simulation studies were conducted based on geometry. The developed computational model is based on the non- 4 MRI datasets of uterine fibroid patients, which were obtained linear Westervelt equation with relaxation effects being taken into during MR-guided HIFU therapy. For each patient, 343 different focus account, bioheat equations for the perfused tissue and blood flow positions within the ablation volume were chosen and the domains. The nonlinear Navier-Stokes equations are employed to de- corresponding phase aberrations for each focus position were obtained scribe the flow in the large blood vessels. A new equation is derived using a stochastic ray tracing method4. Subsequently, 342 positions for the description of large amplitude oscillations of bubbles in visco- served to obtain the new compact base, which was finally validated for elastic medium. The cavitation model is coupled with acoustic and rapid autofocussing. thermal models. RESULTS As Fig. 1 shows, PCA based autofocussing typically re-estab- RESULTS Three dimensional field coupling computational study has lished 90% of the fully phase corrected focus pressure after only 16 been performed. Explicit symplectic finite difference scheme has measurements, whileZernike and in particular Hadamard based cor- been developed for the solution of full wave equation. We used mul- rection require a significantly (5-20 times) longer measurement time, tiple GPUs for the modeling of nonlinear wave equation. to achieve the same improvement. Figure 2 compares the pressures Temperature elevation by focused ultrasound in a minipig has been maps when applying different numbers of PCA base modes and indi- studied experimentally by MRI and numerically. The method for the cates that using about 10 modes are able to reduce the side lobes non-invasive thermal tissue properties has been proposed using MRI. and have similar results as fully phase compensated. Good agreement between the predicted and measured results has CONCLUSIONS Although ARFI based HIFU autofocusing for the com- been obtained. It was also shown that in large blood vessel both pensation of beam aberrations has shown a substantial promise, so Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 99 of 122 far the approach is frequently limited by unacceptably long measure- P78 ment time of several minutes per focus point. Here, we demonstrate A fast 3-D transcranial focused ultrasound simulation based on ray that autofocusing using an individually adapted base derived from tracing 1, 2 2 1,2 prior knowledge can substantially shorten this process. Changzhu Jin , John Snell , Dong-Guk Paeng Ocean System Engineering, Jeju National University, Jeju, Korea; References Focused Ultrasound Foundation, Charlottesville, Virginia, USA [1] Herbert et al., IEEE Trans. Ultrason. Ferroelectr. Freq. Control 56(11) 2388 – 99, Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P78 [2] Larrat et al., IEEE Trans. Ultrason. Ferroelectr. Freq. Control 57(8) 1734 – 7, OBJECTIVES Two objectives are explored: improved skull aberration 2010 correction and focal spot quality visualization. Others have used full [3] Kaye et al., Medical Physics 39 6254 (2012) wave acoustic simulation to explore the possibilities of more accurate [4] Koskela et al., J.Acoust.Soc.Am. 136(3), 2014 phase correction and the simulation and visualization of aspects of focal spot quality [1]. However, the computational cost of such complete acoustic simulations prevents, at the current time, their routine clinical use. In current clinical practice, the location and qual- ity of the hotspot must be evaluated and corrected for during the procedure itself through the use of a number of low-power sonica- tions prior to a final therapeutic sonication [1]. It would bebeneficial to have pre-surgical planning information to inform patient selection and procedure planning in advance of the delivery of energy to a pa- tient. Patient-specific skull characteristics and particular anatomical targets within the skull can make achieving prescribed focal spot temperature elevations difficult or impossible. The focal spot quality in terms of location and temperature could be improved by chan- ging transducer parameters, relative angle between the transducer and patient skull, and morerealistic phase correction. A phase correc- tion taking account of realistic parameters with sufficiently fast com- putation is desired to accomplish these optimizations closer to or within the clinical workflow. In this work, a simplified acoustic simula- tion tool based on GPU-accelerated ray tracing was developed to cal- culate skull aberration correction and to provide real-time visualization of an estimated three-dimensional (3D) pressure field around the targeted focal spot. METHODS A modular 3D planning software system was developed Fig. 1 (abstract P77). The focus pressure versus number of base for transcranial focused ultrasound procedure planning. Within this mode used curves for PCA base, ZP base and H base. Note the rapid environment, CT and MRimages can be displayed and registered. convergence of the PCA acceleratedmeasurement process, which Skull surfaces in the CT image are characterized using a threshold- only takes 4 base modes while ZP and H based process requires based algorithm and a 3D edge detection operator. A single layer much more modes skull model was assumed. A virtual FUS transducer which has a 30cm diameter hemispherical shape and consists of 1024 Fibonacci pat- terned elements was designed. Each transducer element acoustic beam is represented by a computed path through the cooling water, skull and brain tissue. This path from the transducer to the focal point was divided into 3 individual ray segments (water, bone, and brain). The refraction caused by impedance difference at skull bone surfaces was computed based on Snell’s Law. Any ray with an inci- dent angle at the outer skull surface greater than critical angle was neglected. The computation volume of pressure field contains 21×21×21 cubic sample points and the distance between each point on XYZ axis is 1mm. The center of the computation volume was set to lie on the target point. The projection vector of each sample point onto the third ray segment was computed and the length projection was taken as new third ray segment. In order to get a focused spot, the phase at the end of the ray trace was collected and utilized for phase correction prior to pressure computation. The pressure field contribu- tion from each ray trace to the sample point was accumulated using a simplified beam profile model. The attenuation coefficient of water, skull, and brain was set as 0.0022, 0.3103 and 0.0690 dB/cm/MHz re- spectively. The density of the water, skull, and brain was set as 1000, Fig. 2 (abstract P77). Simulated pressure maps obtained after 1x4, 1900 and 1030kg/m^3. The assumption of exponential decay on the 5x4, 10x4, and 256x4 measurements of the autofocusing process. pressure along longitudinal direction was applied and a sinc function Note the pressure map acquired after 10x4 measurements is largely was implemented as pressure decay on the axial direction of the beam. the same as fully phase compensated one, indicating a rapid RESULTS This tool provides an intuitive 3-D view of pre-treatment refocusing process imaging data and also improves the understanding of the focal Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 100 of 122 OBJECTIVES To investigatethedifferentvasculareffectofrabbitVX2 tumor induced by diagnostic ultrasound combined with microbubbles (MB). METHODS Forty rabbit with VX2 tumor were treated by diagnostic ultra- sound with different strengths of Mechanical index (MI) 0.3,0.7,1.4 with MB as experiment group (n=10, n=10, n=10) and negative control (n=10) for 5 minutes. 0.1ml microbubbles in 5 ml normal saline was used as a therapeutic dose. The tumor blood perfusion was imaged with contrast- enhanced US (CEUS) before and after treatment which were analyzed by the contrast analysis software of VINNO 70 . After the treatment, the tumor samples were collected for pathological examination. RESULTS The PI in MI=0.3 increased after treatment, the PI and Fig. 1 (abstract P78). (a) The visualization of wave beam path AUC of MI=1.4 decreased after treatment (Fig. 1). using ray tracing method. (b) The Sagittal plane of simulated CONCLUSIONS Low-intensity diagnostic ultrasound (MI=0.3) com- focal pressure bined with microbubbles can enrich the blood perfusion of VX2 tumor in rabbits, and high-intensity diagnostic ultrasound (MI=1.4) can decrease the blood perfusion. quality on various FUS transducer setups interactively. The pressure field which takes account of attenuation and transmission loss was calculated and displayed within 1 Hz computation speed (Fig. 1b). CONCLUSIONS A fast 3D patient data visualization and acoustic simu- lation software based on ray tracing method were developed (Fig. 1a). The phase correction andtransducer setup could be exported to a third party full-wave simulation software for more accurate prediction of the pressure field. It offers help to the clinician with patient selection and to decide the optimum transducer setup during pretreatment planning. Validation of the accuracy of the pressure field com- puted by this method as compared with full wave simulation methods and hydrophone data is the subject of future work. Reference [1] Clement, G.T. and K. Hynynen, Phys Med Biol, 2002. 47(8): p. 1219-36. P79 Efficacy of ultrasound mediated microbubbles in diclofenac gel to Fig. 1 (abstract P80). Ultrasound contrast image of VINNO70 before enhance transdermal permeation in rheumatoid arthritis induced and after the treatment.(A, B, C, D on behalf of MI=0.3,0.7,1.4 group rat 1 2 and control group, 1, 2 and 3 respectively represent the twodimensional Ai-Ho Liao , Ho-Chiao Chuang ultrasound,ultrasound contrast image before treatment and after National Taiwan University of Science and Technology, Taipei, Taiwan; treatment) Images show that thegroup of MI=0.3combined with National Taipei University of Technology, Taipei, Taiwan microbubbles can enrich the blood perfusion of VX2 tumor in rabbits, Correspondence: Ai-Ho Liao MI=1.4 group decrease the blood perfusion,and there was no significant Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P79 change in the blood perfusion of the MI=0.7group and control group before and after treatment This abstract is not included as it has already been published: Liao A.H, Chuang H.C, Chung H.Y. Efficacy of ultrasound mediated microbubbles in diclofenac gel to enhance transdermal permeation in rheumatoid arthritis induced rat. IEEE. 2015. Available from: P81 http://ieeexplore.ieee.org/document/7319152/. Augmentation of muscle perfusion by microbubble mediated ultrasonic cavitation W. Gao, C. Yi, X. Qiao, Z. Liu Department of Ultrasound, Xinqiao Hospital, Third Military Mediical University, Chongqing, China P80 Correspondence: W. Gao Vascular effect of rabbit VX2 tumour induced by diagnostic Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P81 ultrasound with microbubbles Xueyan Qiao, Zhong Chen, Cuo Yi, Wenhong Gao, Shunji Gao, OBJECTIVES Ultrasound has been known capable of enhancing Zheng Liu tissue perfusion through both thermal and non- thermal bio- Ultrasound Department, Xinqiao Hospital, Third Military Medical effects. The main purpose of this study is to investigate if the en- University, Chongqing, China hanced non-thermal ultrasonic bioeffect- microbubble mediated Correspondence: Xueyan Qiao cavitation could effectively agitate blood perfusion of mice skel- Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P80 etal muscle. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 101 of 122 METHODS Twelve healthy nude rice were intravenously injected microbubbles for both ultrasonic cavitation and contrast imaging. Firstly, one randomly chose posterior limb underwent microbubble mediated cavitation, while the other side went through no treatment as an own control. Intermittent ultrasound applied for cavitation was set at a frequency of 4 MHz, pulse length of 18 cycle, pulse repetition frequency of 50 Hz, mechanical index of 1.4, work/rest interval of 0.48s/2s, and total irradiating time of 10 minutes. Microbubbles for cavitation with a number of approximately 8x10 was injected slowly meanwhile irritate ultrasound was applied. Then simultaneous contrast imaging of both limbs through a transverse section was conducted right after cavitation procedure (Fig. 1). Dynamic contrast video was recorded for evaluating of skeletal muscle perfusion, and quantitative parameters including peak intensity, area under curve, and ascending slope were analyzed. Average values of treatment side and control side were statistically compared using paired sample T test. RESULTS Contrast imaging records exhibited prominent acceleration and abundance of treated muscle perfusion. Entry of microbubbles into peripheral circulation is much more rapid and numerous in the treated side than in the control side (Fig. 2). Quantitative analysis re- sults also showed a steady rise of perfusion on the treated side with statistical significance. Average values of peak intensity, area under curve and ascending slope in control side were (71.2467±10.86445), (6436.7658±981.94242), and (0.7342±0.30417), while in treated side they were (82.0100±11.23804), (7584.8033±916.03143), and (0.9208 ±0.39867),respectively (P<0.05) (Fig. 3). CONCLUSIONS Microbubble cavitation irritated by high mechanical, low duty cycle intermittent ultrasound could effectively agitate mice skeletal muscle perfusion, leading significant raise of both blood vel- ocity and blood volume. Fig. 1 (abstract P81). Illustration of simultaneous contrast imaging of both limbs through a transverse section Fig. 3 (abstract P81). Histogram of peak intensity value, area under curve and ascending slope in the two groups P82 Relationship between different molecule sizes and delivery efficiency of pancreatic cancer cells mediated by different cavitation dose 2,1 1 2 2 3 Lizhou Lin , Mouwen Cheng , Fan Li , Lianfang Du , Alfred C. Yu , Peng Qin Department of Instrument Science and Engineering, Shanghai Jiao Tong University, Shanghai, China; Department of Ultrasound, Shanghai First People’sHospital, Shanghai Jiao Tong University, Shanghai, China, Fig. 2 (abstract P81). Continuous screen shots of contrast imaging 3Department of Electrical and Computer Engineering, The University of after treatment procedure. The treated side (left) showed much waterloo, Waterloo, Alberta, Canada more rapid and abundant perfusion of peripheral circulation than Correspondence: Lizhou Lin the control side (right) Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P82 Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 102 of 122 OBJECTIVES Present studies have validated acoustic cavitation, trig- gered by ultrasound and microbubbles, could enhance the membrane permeability of pancreatictumor cells for improving macromolecular drug delivery. However, the efficiency of different molecular sizes inter- nalized into the pancreatic tumor after acoustic cavitation is not clear. This work aims to address the relationship between different molecule sizes and delivery efficiency of pancreatic cancer cells under different typesand dosesof acousticcavitation. METHODS The pancreatic cancer Panc-1 cells mixed with 1% SonoVue microbubbles and FITC-dextran with different molecule sizes (4 kDa, 40 kDa and 500 kDa) were placed in a tissue mimicking chamber and ex- posed to 1-MHz ultrasound with 1 kHz pulse repetition frequency, 300 cycles and different peak negative pressure (0.25 MPa, 0.6 MPa, 1.25 MPa) (Fig. 1). Another 7.5-MHz focused transducer was employed to passively receive acoustic signals (Fig. 2). The intensities of the ultrahar- Fig. 2 (abstract P82). Panc-1 cells mixed with 1% SonoVue bubbles monics and broadband components were respectively measured to (black water) and water (as control, redcurve) exposed to different quantify the dose of stable and inertial cavitation, respectively. The PNP (0.25 MPa and 1.25 MPa). 1 KHz PRP and 30% duty cycle. samples were stained by Propidium Iodide 30 min after sonication, and Frequency domain of the recorded signal in a period with 0.25 MPa then were analyzed by flow cytometry to assess delivery efficiency and (2-1) and 1.25 MPa PNP (2-3). Power of untraharmonics (2-3) and cell viability. Five replicates experiments were conducted to calculate broadband components (2-4) as a function of exposure time the mean standard deviation and statistically analyze. RESULTS 1. The delivery efficiency (5.90±1.37%) for different FITC- dextran and the necrotic ratio (4.44±1.05%) were not enhanced after Panc-1 cells underwent stable cavitation, compared with the control without exposure to ultrasound (3.41±1.37% and3.08±0.51% for delivery and necrosis ratio, respectively).2. No significant difference (P > 0.05) among the delivery efficiency of FITC-dextran with differ- ent mass was observed when Panc-1 cells were exposed to different inertial cavitation dose (ICD) (Fig. 3). Relatively high ICD induced 30.63±7.73%, 27.57±6.59% and 26.11±5.62% delivery ratio for 4 kDa, Fig. 3 (abstract P82). Delivery efficiency of different FITC-dextran 40 kDa and 500 kDa FITCdextran, respectively.3. Both the delivery (3-1: without ultrasound exposure; 3-2: 4 kDa; 3-3: 40 kDa; 3-4:500 kDa) efficiency for all FITC-dextran and the necrotic cells were positively and necrosis analyzed by flow cytometry Pan-1 cells mixed with 1% correlated with ICD (delivery and necrosis ratio increased by approxi- SonoVue bubbles and FITC-dextran with different molecule size mate 40%and 60%, respectively when ICD increased about 8 times). (4 kDa, 40 kDa and 500 kDa) expsoed to high inertial cavitation dose, The number of the internalized FITC-dextran molecules exhibited corresponding to 1 MHz untrasound 1.25MPa PRP, 1 KHz PRP and negative correlation with molecule size (The fluorescence intensity of 300 cycles the internalized 4 kDa FITC-dextran was almost 110-fold high than that of 500 kDa FITC-dextran, respectively) (Fig. 4). CONCLUSIONS Stable cavitation could not enhance the internalization of macromolecule into pancreatic cancer cells, while the delivery ratio of different molecule sizes was positively correlated with the inertial cavitation dose. Although the delivery rate induced by inertial cavita- tion was not dependent on the macromolecule size below 35 nm, the number of internalized molecules was negatively correlated with the molecule size. We believe these results would provide useful informa- tion for pancreatic tumor therapy mediated by acoustic cavitation. Fig. 4 (abstract P82). Delivery efficiency (4-1), necrosis ration (4-2) and the relative fluorescence intensity of internalized FITC-dextran analysis when the Panc-1 cells mixed with 1% SonoVue bubbles and FITC-dextran with different molecule size (4 kDa, 40 kDa and 500 kDa) exposed to different intertial cavitation doses (***, p<0.001; NS p<0.05) P83 Electron microscopy of renal boiling histotripsy lesions created in an in vivo porcine model 1 2 3 1 1 Y. Wang , S. Buravkov , V. Chernikov , T. D. Khokhlova , G. R. Schade , 1 1 1, 2 W. Kreider , A. Maxwell1, M. Bailey , V. Khokhlova 1 2 University of Washington, Seattle, Washington, USA; M.V. Lomonosov Moscow State University, Moscow, Russian Federation; Research Institute of Human Morphology, Moscow, Russian Federation Correspondence: Y. Wang Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P83 Fig. 1 (abstract P82). Experimental Setup for ultrasound exposure OBJECTIVES Boiling histotripsy (BH) uses millisecond-long pulses of and cavitation detection focused ultrasound (FUS) waves with shocks to mechanically Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 103 of 122 homogenize tissue. Histological evaluation has demonstrated that le- P84 sions with microscopically fine borders can be generated. Here we Subcellular localization of sonosensitizer for autophagy report a preliminary evaluation of the ultrastructural characteristics of cooperated apoptosis in sonodynamic therapy renal BH lesions created in an in vivo porcine model. L. Song METHODS Pigs were treated with transcutaneous BH targeting BME, The Hong Kong Polytechnic University, Hong Kong, China volumes of both kidneys using a 1.5 MHz FUS transducer operat- Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P84 ing at peak electric power ranging from 0.6 – 4kWunder coaxial B-mode ultrasound guidance. Sonication protocols delivered 10 OBJECTIVES Sonodynamic therapy (SDT), based on the synergis- pulses of 1-10 ms duration and 1% duty factor to 4 adjacent tic effect of low-intensity ultrasound and sonosensitizer, is a po- focal points spaced 1 mm apart without respiratory gating. Eu- tential noninvasive approach for the treatment of cancers. thanasia was performed immediately after treatment, and the However, the specific cellular death mechanisms of sonosensiti- treated regions of the kidney were carefully removed en bloc zers with different subcellular localization patterns remains un- and placed in ½ strength Karnovsky’s fixative. The tissue was known. In the present study, protoporphyrin IX (PpIX), either sliced into 1 mm sections parallel to the direction of treatment. endogenously induced by 5-aminolevulinic acid (ALA) or ex- Small regions from the identified lesions and their border ogenously administered, were used to investigate the effects of (1x2mm) were sampled and processed for transmission electron its subcellular localization pattern on key cellular activities in- microscopy (TEM). The remaining tissue slices were processed for cluding mitochondrial dynamics and cargo-selective pathways of histological evaluation. Semi-thin sections were taken from the autophagy in hela cells. TEM blocks and stained with 1% methylene blue to determine METHODS Different concentration and distribution patterns of from which locations to take the ultrathin sections. Ultrathin sec- ALA-induced PpIX and exogenous PpIX in Hela cells were deter- tions were double stained with uranyl acetate and lead citrate mined by microplate reader and confocal immunofluorescence and ultrastructural examinations were performed on JEM-100CX microscopy respectively. Cell viability, apoptosis, and autophagy and Zeiss Libra-120 transmission electron microscopes. were evaluated by microplate reader and flow cytometry. Mito- RESULTS On histologic assessment, lesions appeared to contain a chondrial dynamics and redox balance were examined by west- slurry of homogenized cellular debris without apparent cellular ern blot and confocal immunofluorescence microscopy. cargo- structures, areas of petechial hemorrhage, and a distinct lesion selective pathways of autophagy were determined by confocal border between treated and untreated kidney. At the ultrastruc- immunofluorescence microscopy. tural level, a border region between treated and untreated kidney RESULTS We demonstrated that both autophagy and apoptosis was observed measuring 10-20 microns, in which cell membranes existed in ALA-induced PpIX and exogenous PpIX mediated had been disrupted but intracellular organelles remained intact sonodynamic therapy. However, ALA-SDT mainly caused (Fig. 1). Within the lesion, ultrastructural analysis revealed regions mitochondria-specific autophagy by harming mitochondrial fu- of complete loss of structure with replacement of cells and or- sion and fission balance, exogenous PpIX-SDT caused non- ganelles by electron dense sub-micron cellular debris alternating selective autophagy by harming cytoplasmic proteins and or- with small areas of completely disrupted cells containing organ- ganelles. Furthermore, ALA-SDT caused unbalance of mitochon- elle ghosts. Throughout the lesions, fragmented collagen fibrils drial redox system by excessive production of mitoROS, while were observed (Fig. 1). Within lesions intact erythrocytes were exogenous PpIX-SDT caused unbalance of cytoplasmic redox observed within the slurry of cellular debris (Fig. 1) consistent system by excessive production of cytoplasmic ROS. Recipro- with post-treatment petechial hemorrhage. cally, MnTMPyP, mitoROS scavenger, rescued both cellular au- CONCLUSIONS This data represents the first ultrastructural study tophagy and apoptosis in Hela cells treated with ALA-mediated of BH lesions generated in an in vivo animal model. TEM sonodynamic therapy. evaluation revealed that aside from the presence of intact CONCLUSIONS Taken together, we conclude that different subcellu- erythrocytes, the lesion contents were similar in characteristic lar localization patterns of PpIX induce different cargo-selective path- to that observed in ex vivo tissue. At the border of the lesion ways of autophagy, serving as a pro-death function, cooperated with there is a gradation in ultrastructural changes not observed apoptosis to dictate the fate of cell death in sonodynamic therapy. microscopically. Funding: NIH R01 EB7643, K01 EB015745, NSBRI through NASA NCC 9-58, RFBR 16- 02-00653, and Ur- ology Care Foundation. P85 Focused ultrasound reprograms ethanol-treated prostate cancer cells back to normal Heng Yu, Hakm Murad, Daishen Luo, Damir Khismatullin Biomedical Engineering, Tulane University, New Orleans, Louisiana, USA Correspondence: Heng Yu Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P85 OBJECTIVES Prostate cancer is the most common and sixth deadli- est cancer in men worldwide. Since the majority of prostate cancer patients are elderly men, often not suitable for invasive procedures, there is a need for minimally invasive therapies (e.g., focused ultra- sound) against prostate cancer. Previous ex vivo and in vitro studies Fig. 1 (abstract P83). Representative TEM micrographs from the conducted in our laboratory showed that high-intensity focused center of the lesion (left), at the border of the lesion (center) and ultrasound (HIFU) synergistically enhanced tumor destruction by lower magnification view of the lesion border (right). The lesion ethanol injection. The objective of this study was to investigate the contained intact erythrocytes (E) in a slurry of cellular debris, molecular mechanisms behind anti-cancer effects of HIFU. Specific- fragmented collagen fibrils (arrowhead), and some ghosts of organelles ally, we tested the hypothesis that focused ultrasound drastically re- (G). At the border, intact but damaged organelles (O) lie in between duced the metastatic potential of chemically-treated cancer cells via cellular debris (D) and intact cells (C). At lower magnification (right), the overproduction of heat-shock protein 70 (HSP70), death receptor border region between fully intact and fully fragmented tissue is clearly Fas, its ligand FasL and TNF-α receptor. observed (yellow dotted lines) and measures approximately 20 METHODS DU-145 and PC-3 human prostate cancer cells were microns (black arrow) cultured in T-175 flask in full growth medium. The suspension of Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 104 of 122 cultured cancer cells (100 μl, 2.7million cells/ml) was placed in a P86 0.2 ml thin-wall PCR tube. The cells were left untreated (control) Therapeutic effect of focused ultrasound combined with dendritic or exposed to HIFU alone, 4% ethanol, or HIFU + 4% ethanol. cell treatment for melanoma: preliminary study The focused ultrasound signal was generated by a 1.1 MHz trans- Eun-Joo Park, Yun Deok Ahn, Yuri Cheon, Jae Young Lee ducer in the continuous or pulsed mode, with acoustic power Radiology, Seoul National University Hospital, Seoul, Korea ranged from 4.1 W to 20.52 W. HIFU level 4(H4) in this experi- Correspondence: Eun-Joo Park ment has the acoustic power of 8.72 W. The Fas, FasL and TNF-α Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P86 expression in the cancer cells was measured by flow cytometry at 2, 24, and 72 hours post-treatment. The Hsp70 protein levels OBJECTIVES As focused ultrasound (FUS) combined with microbubbles were determined by Western blot analysis at 2, 12, 24, 48 and 72 has been widely studied in cancer treatment, there is growing interests hours post-treatment. To confirm that cancer cells treated with in immunotherapy combined with FUS as FUS treatment might trigger ethanol and then HIFU lose their aggressiveness, we conducted a the immune response so that it results in therapeutic effects for cancer. series of adhesion and spheroid culture experiments with treated To investigate the effects of FUS on the body’s immune system, this cancer cells. Specifically, we measured the number of cancer cells study was designed as a preliminary study that evaluates the thera- adhered to TNF-α-activated endothelium under static conditions peutic effects of antigen-pulsed dendritic cells (DC) with FUS treatment. and tested the ability of the cells to form multi-cellular spheroids METHODS A total of 30 mice were inoculated with B16-F10 melanoma cells using a hanging drop method. The data (mean ± SEM) are the CFPAC-1 and divided into five treatment groups: control, antigen-pulsed DC result of 3-4 independent experiments. Statistical significance was only (DC), FUS only (FUS), and DC with FUS at two different FUS operating determined by Student’st-test. conditions DC-FUS1, DC-FUS2). Animals were treated on a weekly basis for RESULTS Prostate cancer cells, treated or not with ethanol, signifi- three weeks and posttreatment monitoring was followed for one week. cantly increased their expression of FasL immediately after being ex- RESULTS Animals in the group treated with DC and FUS2 which has posed to HIFU and continued to produce this molecule at a higher mechanical index (MI) and short duty cycle (5%) showed slower significantly higher amount than untreated (control) or ethanol tumor growth incomparison to control and DC only groups. Tumor alone-treated cells at 24 hours and 72 hours post-treatment (Fig. 1A). growth in the group treated with FUS only also showed lower growth The highest expression of FasL was achieved for the combined treat- rate than DC only and control groups. However, the group treated with ment group. Similarly, the combined treatment led to significantly DC and FUS1 which has lower MI and long duty cycle (50%) showed fas- higher expression of Fas than any other treatment at both 24 and 72 ter tumor growth than DC only, DC and FUS2, and FUS2 only groups. hours (Fig. 1B). The expression of TNF-α receptor was also signifi- CONCLUSIONS Based on the result that FUS treatment with high MI cantly increased in treatment group at both 24 and 72 hours (Fig. and very short duration can enhance therapeutic effects of DC treat- 1C). HSP70 expressed significantly higher in cancer cells exposed to ment, it is assumed that mechanical effects of FUS might be the main HIFU than that in untreated or ethanol alone-treated cells (Fig. 1D). mechanism for enhanced treatment effects of DC therapy. In order to Our static adhesion assay showed that the cancer cells in the com- investigate the detail of FUS effects on different therapeutic outcome bined treatment group attached much rarely to TNF-α-activated vas- and to improve the treatment protocol for enhanced therapeutic ef- cular endothelium than the cells in other groups. The cells exposed fects of the combined treatment further study will be followed. to both ethanol and HIFU were unable to form three-dimensional tumor spheroids. CONCLUSIONS Although Hsp70 plays a key role in cancer initiation and progression, its overproduction interferes with NF-κB signaling, P87 thereby causing apoptosis and reduced expression of adhesion mole- The long-term fate of the sonoporated pancreatic cancer cells is cules required for metastasis. Here, we showed that focused ultra- uncorrelated with the degree of the sonoporation sound induces Hsp70 overproduction in prostate cancer cells and 2,1 1 3 2 1 Lizhou Lin , Caixia Jia , Alfred C. Yu , Lianfang Du , Peng Qin promotes cell apoptosis via an increase in expression of Fas, FasL Department of Instrument Science and Engineering, Shanghai Jiao and TNF-α receptor. All these factors lead to phenotypic changes in Tong University, Shanghai, China; Department of Ultrasound, Shanghai surviving cancer cells that reduce their aggressiveness. First People’s Hospital, Shanghai Jiao Tong University, Shanghai, China; Department of Electrical and Computer Engineering, The University of waterloo, Waterloo, Alberta, Canada Correspondence: Lizhou Lin Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P87 OBJECTIVES Sonoporation, refers to the reversible membrane perfor- ation induced by the acoustic cavitation, has showed great potential for macromolecule delivery. Previous studies have determined the fate trend of the sonoporated cell. However, it is commonly believed that the sono- poration appeared to be heterogeneous due to discrete cavitation events adjacent to every single cell. The relationship between the long-term fate trend and the degree of the sonoporated cells is not still clear. This work aims to revealthe long-term fate of heterogeneously sonoporated cells. METHODS The pancreatic cancer Panc-1 cells mixed with 1% SonoVue microbubbles and 40 KDa FITC-dextran were placed in a tissue mimick- ing chamber and exposed to 1-MHz ultrasound with 1 kHz pulse repeti- tion frequency, 300 cycles and 0.6 MPa, and peak negative pressure (Fig. 1). Another 7.5MHz focused transducer was employed to passively detect the inertial cavitation dose (Fig. 2). The samples were firstly stained by Propidium Iodide at 30 min after exposure. After the sono- porated cells were identified by flow cytometry analysis, according to the relative fluorescence intensity (weak, medium and high) of the in- Fig. 1 (abstract P85). A: FasL expression in prostate cancer cells at ternalized FITC-dextran, the sonoporated cells were categorizedinto 2, 24 and 72 hours post treatment. B: Fas (CD95) expression in three sub-groups by flow cytometry sorting. After cultured for 6 H and prostate cancer cells at 2, 24 and 72 hours. C: TNFα expression in 24 H and stained by Annex V-alex350 and Propidium Iodide, these prostate cancer cells at 2, 24 and 72 hours post treatment. D: HSP70 three sub-groups ofsonoporated cells were detected to determine the expression in prostate cancer cell at 2 and 24 hours apoptotic and necrotic ratio by flow cytometry analysis. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 105 of 122 RESULTS The sonoporated cells were categorized into three sub-groups (sub-groups 1-3), which represented different degrees of sonoporation re- spectively (Fig. 3). The fluorescence intensity of the internalized FITC-dex- tran in sub-groups 2 and 3 was approximate 5.62-fold and 19.53-fold higher than that in sub-group 1, respectively. The apoptotic and necrotic ratio in all three sub-groups of the sonoporated cells gradually increased with the increasing culture time, compared with those in the control cells (Figs. 4 and 5). No significant difference in both the apoptotic (P > 0.05) and necrotic (P > 0.05) ratio were observed in three sub-groups of sonoporated cells which were cultured for 6 h and 24 h culture, respect- ively (at 6 H post-exposure, 2.48±1.65%,5.21±1.01 %, 5.05± 1.17 % and 5.95± 1.04 % apoptosis in the control and sub-groups 1-3 sonoporated cells, respectively. At 24 hour post-exposure, 4.40+2.81%,11.15± 2.48 %, 11.17± 4.93 % and 13.21± 3.78 % apoptosis in the control and sub- groups 1-3 sonoporated cells, respectively). CONCLUSIONS Our results indicated some of the sonoporated cells would experience apoptosis and necrosis in the long-term after inertial cavitation. The apoptotic and necrotic ratio in the sonoporated cells exhibited no sig- nificant correlation with the degree of sonoporation. These findings prelimin- arily suggest that the signal, that triggers the apoptosis and necrosis of the sonoporated cells, may be not correlated with the physical damage caused by acoustic cavitation, but dependent on underlying cellular response. Fig. 4 (abstract P87). Three sub-groups (1-3) of the sonoporated cells were respectively cultured for 24 H after inertial cavitation, then stained by Annexin V-Alex 350 and PI, and detected by flow cytometry analysis (A) Viable cells, (B) Sub-group 1 sonoporated cells; (C) Sub-group 2 sonoporated cells; (D) Sub-group 3 sonoporated cells Fig. 1 (abstract P87). Experimetnal Setupfor ultrasound exposure and cavitation detection Fig. 2 (abstract P87). Panc-1 cells mixed with 1% SonoVue bubbles (black water) and water (as control, red curve) exposed to 0.6 MPa PNP, 1 KHz PRP and 30% duty cycle. Frequency domain of the recorded signal in a period (2-1). Power of broadband components Fig. 5 (abstract P87). The apoptosis ratio of three sub-groups as a funtion of exposure time (2-2) sonoporated cells at 6 H and 24 H after ultrasound exposure P88 Tissue erosion at the fluid interface by dual-frequency HIFU excitation Yisheng Lei, Yufeng Zhou Nanyang Technological University, Singapore, Singapore Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P88 Fig. 3 (abstract P87). The sonoporated cells were catergorized into OBJECTIVES High-intensity focused ultrasound (HIFU) burst has been three sub-groups (1-3) according to the fluorescence of the used to successfully erode the tissue or gel phantom at the interface internalized FITC-dextran. A. the fluorescence of the control; B. The of fluid. The performance of histotripsy or microtripsy [high peak fluorescence of the reversible sonoporated cells. C. the relative FITC negative pressure, p-, and pulse repetition frequency (PRF > 100 Hz) of the three sub-groups 1-3 but short pulse duration (in the order of ms)] and those conventional Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 106 of 122 HIFU bursts [moderate p-, long pulse duration (in the order of ms), CONCLUSIONS The results of this study suggest that histotripsy has and low PRF (a few Hz)] have been confirmed. In order to further potential as a completely non-invasive liver cancer ablation method. understand the mechanism of tissue erosion and improve the cap- Future work is ongoing to investigate long-term safety and biological ability, a new technology, using dual-frequency excitations, was pro- response to histotripsy in relevant in vivo liver cancer models. posed and investigated in this study. METHODS The effect of frequency difference and modulation depth in the dual-frequency excitation on the produced erosion area and volume were quantitatively evaluated and compared to those of sin- gle-frequency excitation. In addition, the outcomes at the different PRFs were also compared. Bubble dynamics using different excitation strategies were captured by high-speed photography and passive cavitation detection (PCD). The acoustic field of dual-frequency exci- tation was simulated numerically and measured experimentally. The corresponding bubble oscillation was also simulated with the Gil- more model. RESULTS Dual-frequency excitation is more effective to disintegrate the gel phantom and tissue than single-frequency excitation up to about 2 fold. CONCLUSIONS In summary, this strategy can enhance the tissue ero- sion using the same output power, and operational parameters should be optimized for the bestoutcome. In vivo experiment will be carried out for the clinical translation. P89 Fig. 1 (abstract P89). (A) A custombuilt small animal system with a Non-invasive liver cancer ablation using histotripsy in an in vivo (B) 1 MHz therapy transducer was used to apply histotripsy to Hep3B murine model tumors in an in vivo murine model. During treatment, the (C) bubble Eli Vlaisavljevich, Tyler Gerhardson, Joan Greve, Shan Wan, Kim Ives, cloud and (D) tissue fractionation was visualized in realtime. After Timothy Hall, Theodore Welling, Zhen Xu treatment, histological analysis showed precise lesions generatedinside University of Michigan, Ann Arbor, Michigan, United States the tumor Correspondence: Eli Vlaisavljevich Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P89 OBJECTIVES Current liver cancer ablation methods are primarily ther- mal-based and thus share inherent limitations such as inconsistent ablation in tissue with nonuniform heat dissipation patterns or in- P90 complete tumor necrosis near major vessels. Recently, our group has Comparison of ultrasound-guided high intensity focused developed histotripsy as a completely non-invasive liver cancer abla- ultrasound and surgery for abdominal wall endometriosis: a tion method. Histotripsy is a non-thermal ultrasound ablation retrospective cohort study 2 1 2 method that fractionates tissue through the precise control of acous- Ling Zhao , Jinyun Chen , Chunquan Zhao tic cavitation. Previous experiments in an in vivo porcine model have College of Biomedical Engineering, Chongqing Medical University, shown the feasibility of using histotripsy to noninvasively create well- Chongqing, China; Department of Obstetrics and Gynecology, The 1st confined lesions in the liver through the intact chest, with sharp Affiliated Hospital of Chongqing Medical University, Chongqing, China boundaries between treated and untreated tissue. In this study, the Correspondence: Ling Zhao feasibility of using histotripsy for non-invasive liver cancer ablation Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P90 was tested in an in vivo murine model. METHODS Liver cancer patient-derived xenografts (Hep3B HCC cell OBJECTIVES To compare the safety and efficacy of high-intensity fo- line) were subcutaneously implanted in 8 NSG mice. When tumors cused ultrasound (HIFU) and surgery for abdominal wall endometriosis. reached ~1 cm, histotripsy was applied non-invasively using a custom- METHODS With a retrospective cohort study design, fifty-four Chinese built 1 MHz histotripsy therapy system designed for small animal exper- women with abdominal wall endometriosis who underwent ultrasound- iments (Fig. 1A/B). Guided by real-time ultrasound imaging, histotripsy guided HIFU between January 2012 and December 2014 were enrolled. was applied to the tumors using 1-2 cycle pulses, a pulse repetition fre- In which, 29 cases included in surgery group, 25 cases in HIFU group. quency of 100 Hz, and an estimated in situ peak negative pressure >30 Technological success, adverse events and recurrent rate were assessed MPa.The targeted tumor volume was mechanically scanned using a ro- and compared between two groups. botic micro-positioner controlled by a PC console. After treatment, le- RESULTS The clinical features are comparability, and the success rate sion characteristics were assessed using ultrasound imaging and MRI was obtained of 100% both in the two groups. The clinical efficacy (7T small animal scanner), and the treated tissues were then harvested rate was 92% (22/25) in theHIFU group, and 100 % (29/29) in the sur- for gross analysis and histological evaluation. All procedures were ap- gery group. The numeric rating scales (NRS) after HIFU were signifi- proved by the ICUCA at the University of Michigan. cantly lower than those before the procedure from 6.92 to0.28 RESULTS Histotripsy-induced cavitation clouds were successfully (P<0.05). During the mean follow-up of 32 months (range: 19-46 generated inside the tumors of all subjects, with the bubble clouds months), the duration pain relief were 39.00±30.02 months in the clearly visualized as a hyperechoic zone on ultrasound imaging surgery group and 30.96±28.55 months in the HIFU group (P>0.05). (Fig. 1C). Immediately after treatment, the histotripsy-induced tissue Three cases (10.71%) experienced recurrence of pain in the surgery damage was visualized as a hypoechoic zone on ultrasound imaging group, and 2 (8.00%) in the HIFU group. Adverse events occurred in (Fig. 1D). Gross and histological analysis demonstrated that histotripsy 4 (13.79%) surgery cases and in 1 (4.00%) of HIFU cases respectively, resulted in the complete fractionation of all tumor cells within the tar- without significant difference (P>0.05). Events in the surgery group geted region into an a cellular homogenate with no remaining cellular included incision healing delayed and lung infection, and skin burn structures and sharp boundaries (<10 μm) between treated and un- occurred in the HIFU group. treated tissue (Fig. 1E). No gross damage to the surrounding tissue was CONCLUSIONS HIFU appears to be safe and effective for the treat- observed in any subjects. ment of abdominal wall endometriosis. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 107 of 122 P91 Prediction of thermal lesion formation by short-pulse pre-exposure for cavitation-enhanced ultrasonic heating 1 2 2 1 Ryosuke Iwasaki , Ryo Takagi , Shin Yoshizawa , Shin-ichiro Umemura Biomedical Engineering, Tohoku University, Sendai, Miyagi, Japan; Communications Engineering, Tohoku University, Sendai, Miyagi, Japan Correspondence: Ryosuke Iwasaki Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P91 OBJECTIVES Acoustic cavitation microbubbles are known to enhance the heating effect of ultrasound. In high-intensity focused ultrasound (HIFU) treatment, utilizing microbubbles is expected to accelerate ultrasonic heating to reduce treatment time. However, it is not sim- ple to control the position of generating cavitation bubbles. There- fore, it is necessary to visualize either the bubbles or the effective focal zone where HIFU is most absorbed. If such visualization is per- formed in prior to the cavitation enhanced HIFU treatment, both safety and effectiveness of the treatment will be ensured. The object- ive of this study is to compare the methods between visualizing the ultrasonic backscatter based on B-mode images and the visualizing the ultrasonic attenuation based on acoustic radiation force impulse (ARFI) technique. Fig. 1 (abstract P91). Schematic of experimental set up and HIFU METHODS Figure 1 shows the experimental setup and the sequences sequence as push pulse and therapeutic beam of HIFU exposure for pre-treatment focal zone visualization and tis- sue coagulation. A piezocomposite 2D-array HIFU transducer with both aperture diameter and geometrical focal length of 120 mm was placed in a water tank and connected to a staircase drive amplifier. A sector diagnostic array probe was set in the central hole of the trans- ducer and connected to an ultrasound imaging system. For focal zone visualization, a high-intensity pulse called trigger pulse at an in- tensity of 60 kW/cm2 with a duration of 0.1 ms for generating cavita- tion bubbles was followed by a moderate-intensity HIFU burst at an intensity of 3 kw/cm2 with a duration of 1.9 ms for inducing dis- placements. A chicken breast was used as a sample tissue. Before and immediately after this push pulse exposure, RF data were ac- quired via high-speed ultrasound imaging. The distribution of axial displacements between before and after push pulse exposure was calculated from the phase shift in 1D cross-correlation. After that, tis- sue was coagulated by the repetition of the trigger pulse followed by a HIFU burst with a duration of 44.9 ms, which was continued for 6 s with a duty cycle of 90%. The resulting tissue coagulation was compared with the distribution of the HIFU induced displacement and the B-mode image. RESULTS Figure 2 shows the distributions of HIFU induced displace- ments without and with the trigger pulse, the subtraction B-mode images, and the gross pathologiesof the coagulated tissue. The re- gion of the heat source seems to have been shifted backward from Fig. 2 (abstract P91). Distributions of axial displacements (b), (f), the HIFU focal point. With the trigger pulse, HIFU seems to have subtraction Bmode images (c), (g) and slices of samples after been shielded by the cavitation bubbles. To test this, an offset of coagulation (d), (h) without and with a focal depth offset of 7mm, 7 mm was given to the focal length of the trigger pulse. By respectively, and without trigger pulses following push pulses (a), (e) comparing Fig. 2(b)-(d) and Fig. 2(f)-(h), it is obvious that the off- set extended the area of coagulation deeper, which was well pre- dicted by the effective focal zone by the HIFU induced displacements. The regions of high brightness in the subtraction B-mode images were thought to be the regions where cavitation P92 bubbles were generated. Research on EFT immunity of ashi ultrasonic therapeutic apparatus based on statistical analysis CONCLUSIONS In this study, we proposed HIFU induced ARFI im- Zhiming Zhong, Fangwei Ye, Yang Liu aging employing ultrasonic high-speed imaging to visualize the ef- Biomedical Engineering Collage, Chongqing Medical University, fective focal zone of HIFU. The results showed that the proposed Chongqing, China method was effective for prediction of thermal lesion formation in- Correspondence: Zhiming Zhong duced by HIFU heating even under difficult situation with enhance- Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P92 ment by cavitation bubbles. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 108 of 122 OBJECTIVES Ashi ultrasonic therapeutic instrument can effectively al- CONCLUSIONS Certain intensity and exposure time of ultrasound leviate the chronic soft tissue injury pain, the EFT immunity influ- can enhance the diffusion effect of small beeds solution in the ences the therapeutic effect ofthe equipment. In order to diminish agarose gel. the randomness in the EFT immunity test, it is essential to adopt stat- istical analysis for research on EFT immunity of Ashi ultrasonic thera- peutic apparatus. METHODS The parameters of EFT generator were set according to test standard GB/T 17626.4 or IEC 61000-4-4, but the EFT initial voltage was set as 200V. The power line of Ashi ultrasonic therapeutic apparatus was injected with EFT. Then, the operative mode of equipment was ob- served as a evaluation criteria for judging whether the apparatus was disturbed. If not, the EFT voltage was increased with a step 200V, and continuing test after 30 seconds until that the equipment was dis- turbed. Then, one test was finished and the EFT voltage at this time was record, which was called EFT interference threshold voltage. Ac- cording to the above method, the EFT immunity test was carried out in many times for obtaining multiple EFT interference threshold voltage data which was used to solve unknown parameters in two parameter Weibull distribution model through Maximum Likelihood Method. The distribution law of EFT interference threshold voltage of Ashi was ana- lyzed by theprobability density function (PDF) and cumulative distribu- Fig. 1 (abstract P93). Control sample, no ultrasound applied. The tion function(CDF) curve of Weibull distribution model which has been Xdirection is along the ultrasound beam, the Yaxis is the fluorescent verified by chi-square goodness of fittest. intensity. The profile shows the crosssection the sandwich gel RESULTS According to the result of chi-square goodness of fit test, two parameter Weibull distribution model was available for research on dis- tribution law of EFT interference threshold voltage of Ashi ultrasonic therapeutic instrument. The PDF and CDF curves prove that EFT inter- ference threshold voltage of Ashi equal 900V, and the EFT with differ- ent voltage has different interference probability for Ashi. CONCLUSIONS The test standard GB/T 17626.4 or IEC 61000-4-4 proves that the standard voltage in the EFT immunity test equal 1000V which is the actual output voltage, the value greater than EFT interference threshold voltage of Ashi 900V. Therefore, the apparatus cannot pass EFT immunity test in great probability. The long data communication cable is coupled by EFT easily, which is main reason for Ashi cannot pass test probably. So, it is essential to shorten properly power line or data communication cable in the equipment. The statistical analysis is scientific and useful for obtaining EFT interference threshold voltage of equipment, which has important significance for the research on EFT immunity. Analyzing EFT immunity of equipment based on statistical analysis could decrease effectively randomness in the test result. And the distribution law of EFT interference threshold voltage of instrument will be analyzed accurately and clearly by statistical model. Fig. 2 (abstract P93). Control sample, no ultrasound applied. The Xdirection is along the ultrasound beam, the Yaxis is the fluorescent P93 intensity. The profile shows the crosssection the sandwich gel Experimental study of the promotion of diffusion process in a biofilm by low intensity ultrasound Dong Ma Physics, University of Vermont, Burlington, Vermont, United States Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P93 OBJECTIVES Experimentally to show the Low intensity ultrasound can dramatically enhance nanometer liposome penetration into the biofilm films. METHODS The biofilm model sample was made of three layers of 1% agarose gel; each layer has the thickness of 0.5 mm. The first layer of agarose contains fluorescent beads (diameter: 100 nm) was made in a petri dish, after the agarose was cooled down, a second layer was made on top of the first layer; at this point a two layer agarose gel was made. Flip it over, and made the third layer on top, then we had a three-layer gel with a fluorescent layer sandwiched between two non-fluorescent agarose gels. An ultrasonic tone burst (frequency=2.25 MHz, 10% duty cycle and spatially and temporally averaged intensity, ISATA 4.4 W/cm2) generated by a transducer of diameter 1.9 cm was used to treat the sample for 10 minutes. RESULTS When ultrasound was applied, the diffusion distance Fig. 3 (abstract P93). Ultrasound applied. The Xdirection is along (0.8 mm) is much longer than those without ultrasound (0.1mm). the ultrasound beam, the Yaxis is the fluorescent intensity. The profile Figures 1 and 2 show that there are steep drops on the right side shows the crosssection the sandwich gel (which we are interested in), Figs. 3 and 4 show a very different profile. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 109 of 122 because of the acute pain. The pain threshold was recorded when the muscle contraction at the first time. The pain threshold value in- creasing of ultrasound group were statistically significant compared with the control group on the 3rd and muscle contraction at the first time. The pain threshold value increasing of ultrasound group were statistically significant compared with the control group on the 3rd and 24th (P < 0.05), 7th, 10th and 17th day (P < 0.01) (Fig. 1).Inflam- mation levels: The levels of the 5-HT (Fig. 2A) and PGE-2 (Fig. 2B) in the injured hind leg muscle of rabbit were reduced largely in focused ultrasound irradiation group from the 3rd dayto 17th day after the first treatment (P < 0.05), and the IL-1β level appeared a decreasing trend until the 24th day from the first treatment (P < 0.05) (Fig. 2C). CONCLUSIONS It can be concluded that the focused ultrasound de- vice used in this study is safe and effective in the treatment of soft- tissue injury by cutting inflammation factor expression and increas- ing pain threshold. Fig. 4 (abstract P93). Ultrasound applied The Xdirection is along the ultrasound beam, the Yaxis is the fluorescent intensity. The profile shows the crosssection the sandwich gel P94 The effect of focused ultrasound to treat the soft tissue injury on rabbit models Dandan Liang Chongqing Medical University, Chongqing, China Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P94 OBJECTIVES To evaluate the efficacy of focused ultrasound in the treatment of soft-tissue injury and preliminary discussed mechanism. METHODS Ethics statement: The Ethics Committee of Animal Experi- ments at Chongqing Medical University (Production License No. SCXK 2012-0001) approved the experiment and animal care protocols. Animals: Male and female New Zealand White rabbits (3-4 months old, 2.0–2.5 kg) were chosen from the Animal Center at Chongqing Fig. 1 (abstract P94). See text for description Medical University. All animals were housed in individual cages with a 12:12-h light–dark cycle and a temperature- controlled environ- ment (24 ± 2°C) and given a standard laboratory diet with drinking water at liberty. The animals were adapted to their environment at least 1 week before the experiment started. Soft-tissue injury model construction by hammer blow: As previously reported, the rabbits were fixed in the lateral and outside of the left hind was shaved by shaver and hit thrice on the thigh muscle by a cylindrical hammer (200 g in weight, 2.8 cm in bottom diameter), which fallen freely and vertically from inner of a hard smooth plastic tube (12 cm in length and 3 cm in inner diameter).The act repeated once every two days. When struck for four weeks and bred normally for three days later, the closed soft tissue injury was formed without femoral fractures. Intervention: All model rabbits were grouped into two groups ran- domly:ultrasound treatment group and non ultrasound group. The ultrasound group received a focused ultrasound treatment by a fixed and mobile method, which rabbits were treated fixedly for 20s, then taken a 5s break, lastly treated by mobile method with, 5–10 mm/s for 60s. They were treated per day for continuous ten days. Pain threshold determination: After treatment 30min, the pain threshold was measured by Analgesy-Meter. The rabbit was fixed and pres- Fig. 2 (abstract P94). See text for description sured gradually by the forcing bar and the value, the animal's pain threshold, was recorded when the animal retracted its leg. Measure- ment of PGE2,5-HT and IL-1β concentrations: The muscle 1g was ho- mogenized in 2ml phosphate buffered solution (PBS pH 7.4). Then P95 centrifuged and the supernatant was collected to detect the prosta- Detection of harmonic signal of the focused ultrasound based on glandin-E2(PGE-2),5-hydroxytryptamine(5-HT) and interleukin-1 beta acousto-optic effect (IL-1β) with an enzyme immunoassay at the 1st, 3rd, 7th, 10th, 17th, Fu Li, Hua Wang 24th days after the first of treatment. The parameter was measured Chongqing Medical University, Chongqing, China according to the protocols of respective kits (Shanghai Hushang Bio- Correspondence: Fu Li engineering Institute, China). Statistical analysis: Results were Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P95 expressed as mean ± standard deviation (SD). SPSS 19.0 (SPSS Inc., Chicago, IL, USA) was used for all statistical analyses. Independent sam- OBJECTIVES To explore a non-invasive method for detecting the har- ples t-test was used to analyze all experimental data. P < 0.05 (95% monic signal of high intensity focused ultrasound (HIFU) confidence interval) was considered statistically significant. METHODS When a parallel laser beam passed through the focus of the RESULTS Pain threshold value: When the stimulation intensity was in- focused ultrasonic field, the light signal was converted into an electrical creased gradually, the rabbits would show muscle contraction signal by a photodiode in the distance, and the electrical signal was Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 110 of 122 simulated by the Matlab. To verify the theoretical model, a focused P97 ultrasonic transducer was used as the source of ultrasonic emission, The value of the parameters in the peripheral blood routine test and it was placed in degassed water. A He-Ne laser was used to trans- for the preoperative diagnosis of uterine sarcoma: a review of a mit beam crossing the focus of the sound field, and the optical signal multicentre study in western China whose crossed focus was received by a photodiode (it was connected Dan Li, Wenzhi Chen, Jinyun Chen to the photoelectric detection circuit) in the distance, the output signal College of Biomedical Engineering, Chongqing Medical University, of the photoelectric detection circuit was collected by a digital oscillo- ChongQing, China scope, and the spectrum of the signal was obtained by Matlab (Fig. 1). Correspondence: Dan Li Comparison the normalized frequency spectrum of theoretical and ex- Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P97 perimental obtained. RESULTS The normalized frequency spectrum obtained from the simu- OBJECTIVES The current study was to examine the accuracy of pre- lation of the theoretical model is consistent with the experiment results. operative diagnosis of uterine sarcomas in western China and to evalu- CONCLUSIONS The method proposed in this paper is feasible to de- ate the efficacy of the indicators in the peripheral blood routine test for tect the harmonic signal of HIFU field, and it also lays a theoretical the differential diagnosis between uterine sarcomas and leiomyomas. and experimental foundation tomeasurement the sound pressure of METHODS A total of 102 patients with uterine sarcoma were evaluated, the nonlinear HIFU at focus. underwent surgery in the first affiliated hospital of Chongqing Medical University, the DapingHospital affiliated to the Third Military Medical University, People’s liberation Army of China and the affiliated Hospital of Southwest Medical University, covering from January 1st 2010 to De- cember 1 st 2015. Meanwhile,119 women with leiomyoma, complete clinical and pathological information documented at the time of sur- gery were selected as controls between January 1st 2010 to December 1 st 2015. Study parameters included age at the time of surgery, clinical findings, blood test results, imaging examinations results (specifically ultrasonography and magnetic resonance imaging [MRI]), endometrial cytology findings, postoperative pathological diagnosis and follow-up. Receiver operating characteristics (ROC) analysis was used for specificity and sensitivity estimates. The resulting area under the curve (AUC) indi- cates the average sensitivity of a marker over the entire ROC curve. Fig. 1 (abstract P95). The normalized frequency spectrum obtained Multivariate analysis was performed using nonlinear model of Logistic from the simulation of the theoritical model is consistent with the regression analysis and partial leastsquares-discriminant analysis. experiment results RESULTS At the final preoperative diagnosis, 59.8% (61/102) of uter- ine sarcomas were diagnosed as having malignant disease. 6 indexes including the WBC, neutrophil, monocyte, NLR, MLR and NWR had a certain diagnostic value evaluated by the ROC curve (AUC>0.70) (Fig. P96 1). A comprehensive system combined with by six markers for identi- Effect of cavitation induced by High Intense Focused Ultrasound fication of uterine sarcoma were analyzed by the ROC curve,non- (HIFU) on tungsten filament and stainless steel filament at 10 MPa linear model of Logistic regression analysis and partial least squares- static pressure discriminant analysis (Fig. 2).The AUC of this comprehensive diagno- 1,2 1,2 2 2 2 Yurong Zhang , Faqi Li , Xiaobo Gong , Qi Wang , Guangrong Lei , sis system was 0.90 (95%CI, 0.86to 0.94; sensitivity =76.1%,specificity 1,2 Zhibiao Wang =89.1%).Logistic regression analysis and partial least squares-discrim- State Key Laboratory of Ultrasound Engineering in Medicine Co- inant analysis showed that the comprehensive system combined founded by Chongqing and the Ministry of Science and Technology, with by six markers had a better value of differential diagnosis be- College of BiomedicalEngineering, Chongqing Medical University, tween uterine sarcomas and leiomyomas. Chongqing, China; National Engineering Research Center of Ultrasound CONCLUSIONS The comprehensive system combined with by six Medicine, Chongqing, China markers (WBC, neutrophil, monocyte, NLR, MLR and NWR) in the per- Correspondence: Yurong Zhang ipheral blood routinetest, which is convenient, reproducible, and in- Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P96 expensive, had a potential value of differential diagnosis between uterine sarcomas and leiomyomas (Fig. 3). OBJECTIVES To illustrate the intense cavitation and high energy density induced by spherical cavity transducer with open ends. METHODS We processing on tungsten filament and stainless steel fila- ment with the same diameter based on the extreme transient condi- tion, such as high pressure and high temperature during cavitation induced by continuous high intense focused ultrasound (CHIFU) and pulsed high intense focused ultrasound (PHIFU) at 10 MPa static pres- sures. The images are taken using a high speed camera, and scanning electron microscope (SEM) fractographic analysis is also conducted. RESULTS The 0.2 mm diameter tungsten filament and stainless steel filament were severed by CHIFU (6000 W of electric power) cavitation within 1.39 s and 0.56 s, respectively. PHIFU (1,5000 W of electric power) cavitation could severed 0.2 mm diameter tungsten filament within 0.22 s. SEM fractographic analysis showed that the fracture of stainless steel filament was general fatigue fracture, but the fracture of tungsten filament was relatively complex, which might be in- volved annealing andrecrystallization texture. CONCLUSIONS The tungsten filament is more difficult to be severed by acoustic cavitation than stainless steel filament, and HIFU can effect- ively improve plastic of tungsten filament. Next, we hope to calculate Fig. 1 (abstract P97). Comparison of differential peripheral blood the pressure and temperature of cavitation core region, known as hot- parameters by the receiver operating characteristic curve. AUC>0.70 spot, from the perspective of micro fracture mechanicsto illustrate the was considered to have a certain diagnostic value high energy density of our system. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 111 of 122 METHODS Figure 1 shows the measurement setup of the optical phase contrast method. Ultrasound pressure field in water creates a modulation of the refractive index, which further modulates the phase of the light passing through the field. The non-diffracted com- ponent of the light was focused exactly at the focal point of the sec- ond convex lens, while the other component diffracted due to the phase modulation was laterally slightly away from the focal point. By shifting the phase of the non-diffracted component by the phase plate typically by 90 degrees, the phase modulation of the diffracted component was converted to amplitude modulation through inter- ference with the phase-shifted non-diffracted component and then measured by the camera. The numerical calculation steps based on the measurement were as follows,1, An upstream field in which the optical phase does not wrap and the effect of nonlinear propagation can be ignored was chosen.2, The 3D pressure field was recon- structed from the projected 2D data from measurement by a CT al- gorithm.3, Nonlinear ultrasonic propagation was simulated using the obtained upstream pressure field as the input. In this study, an axi- symmetric 8-element annular array transducer (outer diameter: 80 mm, inner diameter: 36 mm, focal length: 80 mm, center frequency: 1.4 MHz) wasdriven at 65.6Vpp, the pressure field 40 mm upstream the focal point was measured, and the numerical simulation based Fig. 2 (abstract P97). Receiver operating characteristic curve on a pseudo spectral method was performed. The pressure ±5 mm analysis for combined with six markers from the focal point was also measured by scanning a fiber optic hydrophone (Onda, HFO-690) to compare with the proposed method. High-intensity focal field at a therapeutic level can be calcu- lated by multiplying the measured low-intensity pressure before in- putting it to the simulation of nonlinear propagation. Here, it was assumed that the nonlinear propagation at the upstream of the mea- sured area can be ignored even when the acoustic pressure is multiplied. RESULTS Figure 2 (a) shows the optically measured axisymmetric pulsed pressure field upstream the focal point, and Fig. 2 (b)-(e) show the results from the numerical simulation using it as the input. Figure 3 shows the comparison of the obtained focal field between hydro- phone scanning and the proposed method. The pressure distribution in axial and lateral directions is shown. The absolute positive and negative pressures measured by hydrophone were 2.89 and -1.75MPa, whereas they were 2.72 and -2.03MPa determined by the Fig. 3 (abstract P97). Partial least squares-discriminant analysis optical measurement. Good agreement was observed for the wave- (PLS-DA) combined with six markers forms and absolute pressure in both directions. CONCLUSIONS Pulsed ultrasound pressure field was quantified by numerical acoustic holography based on a fast optical measurement using a phase contrast method in combination with a CT algorithm. P98 Both pressure waveform and absolute pressure from the proposed Quantitative measurement of high-intensity pulsed ultrasound method agreed well with that from hydrophone scanning. Further pressure field using combination of optical phase contrast method studies underway are applications to high pressure field at a and acoustic holography therapeutic level and to continuous wave field such as that of HIFU 1 2 2 2 Takuya Nakamura , Hiroki Hanayama , Ryo Takagi , Shin Yoshizawa , (high-intensity focused ultrasound). Shin-ichiro Umemura Biomedical Engineering, Tohoku University, Miyagi, Japan; Communication Engineering, Tohoku University, Miyagi, Japan Correspondence: Takuya Nakamura Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P98 OBJECTIVES In recent years, ultrasound has been widely used for therapeutic as well as diagnostic purposes. To assure safety and effi- cacy of such application, accurate measurement of ultrasonic pres- sure field is necessary. Hydrophone scanning is the most common method to measure ultrasound pressure field. However, this method requires very long scanning time and has the risk of disturbing pres- sure field. In this study, pulsed ultrasound pressure field was quanti- fied by numerical acoustic holography based on a fast optical measurement using a phase contrast method in combination with a Fig. 1 (abstract P98). Optical phase contrast measurement setup CT algorithm. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 112 of 122 intensity and surface temperature is determined by the boundary condition of those two interfaces. Boundary condition of two inter- faces is investigated theoretically in this study. METHODS At the absorber/water interface, acoustic streaming generated by ultrasound lash the interface. In the focal region, the ultrasound field distributed like a cylinder-shape, so in this paper, the flow heat transfer analogy for a single round jet stream. According the time-average energy conservation from the transducer to the focal region, the relationship between average jet velocity and heat transfer coefficient at absorber/water inter- face and focal intensity can be derived, as show in Figs. 1 and 2. At the absorber/air interface, due to the ultrasound irradiation, the surface temperature elevated may cause the natural convection and radiant heat transfer. So the combined heat transfer can be expressed as ΦTotal =ΦConvection + ΦRadiation = AhcΔT+AhrΔT= AhTotalΔT, where htotal is combined surface heat transfer coeffi- cient, hc is convective heat transfer coefficient, hr is radiant heat transfer coefficient, T is the temperature elevation. The relation- Fig. 2 (abstract P98). Optically measured pulsed pressure field ship between transfer coefficient and temperature change is upstream the focal point (a), and the numerically simulated fields (b) shown in Fig. 3. 8.6 µs (c) 17.1 µs (d) 25.7 µs after the input, and(e) at the focus RESULTS According to the boundary analysis, the average heat trans- fer coefficient at absorber/water interface has no significant impact to the absorber/air temperaturechanges, for absorber's thicknesses from 1~4 mm, as shown in Table 1. When the surface temperature changes below 100 °C, the results from combined surface heattrans- fer and heat transfer was closed, the difference of maximum temperature elevation and temperature change rate less than 0.7%, as show in Table 2. CONCLUSIONS According the study, the boundary condition at water/absorber interface can be treated as infinite and constant temperature boundary conditions, when the heating time is very short. And at absorber/air interface, the heat conduction plays a major role, when the interface temperature changed below 100 °C. Fig. 3 (abstract P98). Comparison of pressure waveform between proposed method and hydrophone scanning in axial (a) and lateral (b) directions P99 Boundary analysis of absorber irradiated by the focused ultrasound 1 2 1 1 Ying Yu , Guofeng Shen , Chunlei Lei , Helin Zhang School of Computer, Jiangxi University of Traditional Chinese Medicine, Nanchang, Jiangxi, China; School of Biomedical Engineering, Shanghai Jiao TongUniversity, Shanghai, Shanghai, China Correspondence: Ying Yu Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P99 OBJECTIVES Characterization of focused ultrasound field is important factor for both efficacy and safety of clinical treatment. Recently, the infrared imaging has been used to estimating the intensity of fo- cused acoustic field. The principle of this method is estimating the absolute intensity and relative distribution of focused ultrasound by the temperature elevation at the surface of an absorber which irradi- ated by the focused ultrasound. There were two interfaces, absorber/ air and water/absorber, with large differences in thermal physics Fig. 1 (abstract P99). See text for description properties. In theory, the derived relationship between incident Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 113 of 122 P100 An inverse method for estimating the acoustic intensity in the HIFU field by infrared thermometry 2 1 1 1 Guofeng Shen , Ying Yu , Chunlei Lei , Helin Zhang School of computer, Jiangxi University of Traditional Chinese Medicine, Nanchang, Jiangxi, China; School of Biomedical Engineering, Shanghai Jiao Tong University, Shanghai, Shanghai, China Correspondence: Guofeng Shen Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P100 OBJECTIVES The 3D acoustic field distribution of HIFU is not only the key parameters for evaluating the quality of the transducer, but also an important indicator of the security and efficiency of HIFU system. Recently, a new method which based on infrared (IR) imaging was in- troduced. Authors (A. Shaw, et al and M. R. Myers, et al) have estab- lished the relationship between absorber surface temperature and incident intensity during the absorber was irradiated by the trans- ducer. Theoretically, the shorter irradiating time makes estimation more in line with the actual results. In this study, an inverse method was introduced to estimating the acoustic intensity of HIFUfield using the surface temperature (Fig. 1). METHODS The physical definition of this inverse problem is to recon- struct the unknown incident intensity distribution by measuring the thermal field of the absorber asa function of time. So, the thermal field of the absorber was measured with the IR camera not only dur- ing the heating time but also including the cooling time and pre- heating time. The algorithm for the solution consists of 9 steps.1) Fig. 2 (abstract P99). See text for description choose an initial guess2) solve the direct problem to obtain the sur- face temperature 3) solve the adjoint problem to find the gradient to the function J' 4) compute the conjugate coefficient 5) compute the search direction 6) solve the sensitivity problem with input data 7) compute the step size in the search direction 8) compute the new estimate 9) interrupt the iterative procedure if the stopping criterion is satisfied. Otherwise repeat the iteration until convergence is achieved. RESULTS The method proposed for quantitative assessment of acoustic intensities using IR camera with inverse method has a satis- factory percentage difference, in both maximum intensity (< 13.73 %) and -6 dB beam width (< 10.04 %) in focal region, in comparison to the theoretical simulation using a three-layer medium model (Fig. 2). Fig. 3 (abstract P99). See text for description CONCLUSIONS Thepercentagedifferenceincreasewiththeheat- ing time with zero mean noise, but decrease with heating time when the noise can be ignored. Table 1 (abstract P99). The table of maximum temperature change rate with different average heart transfer coefficient at absorber/ water interface Table 2 (abstract P99). The table of maximum and the maximum temprature change rate with different average heat transfer coefficient at absorber/water interface and heart conduction Fig. 1 (abstract P100). Comparison of the inverse method (red and blue dot line) and theoretical simulation (blue solid line) for heating of 120 and 200 ms Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 114 of 122 Table 1 (abstract P100). See text for description RESULTS Sound field distribution of propagating and standing wave focusing (Fig. 1) There were obvious differences of sound field distri- bution of sound propagation axis between propagating and standing wave focusing. Comparation of nonlinear effects of the two focusing modalities on the same focal pressure (Fig. 2) The nonlinear ef- fect of propagating wave is stronger than that of standing wave because of the stronger distortion of waveform. Recording the formation of lesion in phanton with high-speed camera (Fig. 3) Lesion formation of the propagating wave focusing modalities has been formed the “ellipsoidal” lesion whose major axis was along with wave propagation direction (Fig. 3a). However, lesion formation of focusing of standing wave has been formed the “gourd string” lesion (Fig. 3b). The focal 3-6MHz broadband emis- sion during exposure collected with PCD on the same focusing pressure (Fig. 4) There have been no significant differences of the cavitation signal between the two focusing modalities. CONCLUSIONS 1. Lesion formed in the condition of standing wave focusing is obviously different from that in the condition of propagating wave2. Focusing modality of standing wave need lower power than that of propagating wave to get the same acoustic pressure3. Nonlinear effect plays an important role in the formation of lesion. Fig. 2 (abstract P100). Comparison of the inverse method (blue dot line) and theoretical simulation (red solid line) for heating of 120 ms Table 2 (abstract P100). The difference between the maximum intensity on-axis using the proposed method in the simulated theoretical prediction for different heating time with same noise Fig. 1 (abstract P101). Sound field distribution of three axes of propagating wave focusing (a) and standing wave focusing (b) P101 Comparation of the focusing ultrasound of propagating and standing wave in sound field and bioeffects 2,1 2,1 2,1 2,1 Man Luo , Faqi Li , Zonggui Chen , Yurong Zhang National Engineering Research Center of Ultrasound Medicine, Chongqing, China; State Key Laboratory of Ultrasound Engineering in Medicine Co-Founded by Chongqing and the Ministry of Science and Technology, College of Biomedical Engineering, Chongqing Medical University, Chongqing, China Correspondence: Man Luo Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P101 OBJECTIVES This study aims to investigate the differences of this focus- ing modality with the conventional one in sound field distribution and Fig. 2 (abstract P101). Acoustic pressure waveform of propagating the process of lesion formation under the same acoustic pressure. wave (a) and standing wave (b) on the same focusing pressure METHODS Setting two identical transducers in opposition whose internal (16MPa) surface were on the same spherical surface and driving simultaneously to achieve the focusing of standing wave. And then driving one transducer and sheltering another one from ultrasound beams with sound absorbing materials to get the focusing of propagating wave. The sound field distri- bution of the two focusing modalities were acquired by fiber-optic hydro- phone. The change of the acoustic pressure of focus with the increase of power were measured by fiber-optic hydrophone also and waveform of acoustic pressure were recorded to analyze nonlinear effects of the two fo- cusing modalities on the same focusing pressure (16MPa). Linear fitting of acoustic pressure squared with power extrapolate the powers of getting the experimental acoustic pressure (18MPa). Tissue mimicking phantom were exposed with the extrapolating powers of the two focusing modal- Fig. 3 (abstract P101). Recording the formation of lesion in ities. Meanwhile, the processes of lesion formation in phantom were re- phanton with highspeed camera; (a) the processes of lesion corded with high-speed camera to analyze the differences of lesion formation of the propagating wave focusing modalities; (b) the formation between the two focusing modalities and the focal 3-6MHz processes of lesion formation of focusing of standing wave broadband emissions were recorded with PCD to analyze cavitation. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 115 of 122 value is 0.47, while the values of the erosion treated by histo- tripsy distributed between 0.2 and 1.2 and the meanvalue is 0.40 (Fig. 2). From the results, we find that the mean values of the Nakagami parameter M for the thermal lesion and the mechanic- ally homogenized lesion were different. CONCLUSIONS The preliminary study on the tissue-mimicking phan- tom suggested that the Nakagami parameter M may have the poten- tial possible to identify the lesions treated by HIFU of different modes and realize the structure characterization for the different HIFU lesions in the tissues. Fig. 4 (abstract P101). Cavitation signal of the propagating wave focusing modalities (the black curve) and the standing wave focusing modalities (the green curve) on the same focusingpressure (18MPa) Fig. 1 (abstract P102). The images of the different HIFU lesions. P102 (a) Bmode image of the thermal lesion; (b) Nakagami image of the Feasibility of using nakagami distribution in structure thermal lesion; (c) Bmode image of the lesion byhistotripsy; (d) characterization of the different lesions treated by HIFU Nakagami image of the lesion by histotripsy Meng Han, Na Wang, Mingxi Wan Biomedical Engineering, Xi'an Jiaotong University, Xi'an, Shaanxi, China Correspondence: Meng Han Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P102 OBJECTIVES High-intensity focused ultrasound (HIFU) is a noninvasive technique for tissue ablation and tissue erosion due to the thermal or mechanical effects. However, the microstructure of the treatment target has difference after treated by the different high-intensity focused ultrasound (HIFU) therapy mode. It is important to monitor the thera- peutic effect. The Nakagami image was proposed to better visualize the tissue structure and scatter properties combined with the use of the B-modeimage simultaneously. The aim of this study was to explore the feasibility of using the Nakagami image for structure characterization of the lesions induced by focused ultrasound gradually Fig. 2 (abstract P102). The histograms of the M values of the lesion from the ablation to tissue erosion. area. (a) thermal lesion; (b) lesion by histotripsy METHODS Experiments were conducted on polyacrylamide phan- toms with bovine serum albumin (BSA) using a single-element 1.6-MHz transducer to produce tissue ablation and tissue erosion. After the treatment, ultrasound B-mode images and correspond- ing RF data was recorded and the Nakagami images was pre- P103 sented after the RFdata was processed. B-mode image and Dynamic modelling of piezoelectric hydrophones Nakagami image were combined to visualize the tissue structure Sun Yingqi, Yang Zengtao and scatter properties for structure characterization. Chongqing Medical University, Chongqing, China RESULTS For the B-mode images, the thermal lesion and the his- Correspondence: Sun Yingqi totripsy both appeared to be hyperechoic in the region of the le- Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P103 sion (Fig. 1). In the Nakagami images, the thermal lesion appeared to be homogeneous blue, and has a clear boundary OBJECTIVES The conventional hydrostatic model based on off- with the surrounding tissue. For the lesion treated by the histo- resonance hydrophone design is incapable of analyzing the tripsy, it appeared blue in most of the lesion area, and there is hydrophones operatedaroundandupon the resonance fre- an area of red in the center of the lesion. From the histogram of quencies. To expand the operating limits of the hydrostatic the M values of the lesion areas, we find that the values of the model. thermal lesion concentrated between 0.4 and 0.55 and the mean Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 116 of 122 METHODS A dynamic fluid–structure interaction model for analyz- ing the piezoelectric hydrophones is presented. An analytical so- lution for output voltage of the hydrophone operated in thickness-stretch vibration is derived. RESULTS The output voltage is proportional to the acoustic pres- sure and the piezoelectric constants. The output voltage is also inversely proportional to the elastic stiffness and dielectric con- stants of the material, and the surface area of the hydrophone. Hence, a soft material with a higher-electromechanical coupling factor will produce higher output voltage. However, around the resonance range, the output voltage is seen to strongly depend on the operating frequencies of the hydrophone and the parame- ters of the fluid. CONCLUSIONS The theoretical result shows that the output volt- age obtained by dynamic model in the off-resonance range agrees well with those obtained by hydrostatic model. Further- more, the dynamic model we present is sufficient to analyze the piezoelectric hydrophones over the entire frequency range, espe- cially around the resonance, expanding the operating limits of hydrostatic model. P104 Experimental study of micro-bubble enhance low-frequency and low-intensity ultrasound-mediated plasmid transferring into mycobacterium smegmatis Huimin Zheng Chongqing Medical University, Chongqing, China Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P104 OBJECTIVES To investigate the synergistic effect of microbubble combined with low frequency and low intensity ultrasound (LFLIU) mediated plasmid transfer into mycobacterium. METHODS A mixture of mycobacterium tuberculosis and plasmid were randomly divided into 3 groups: the control group, the ultra- sonic irradiation group and the combination of microbubble and ultrasonic irradiation group. The transformation efficiency, the vitality of the bacteria and the expression of recombinant gene ClpP2 were measured respectively by colony-counting methods, flow cytometry methods, the Real-time quantitative polymerase chain reaction after the experiment processing. RESULTS Positive colonies were selected on the LB agarose plate containing kanamycin after ultrasound exposure, the efficiency of the combination of microbubble andultrasonic irradiation group was 19.33×102CFU/ngDNA, which was more 19 times effi- cient than the ultrasonic group. The survival rate of the bacteria of ultrasonic irradiation group was 75.76%, which was more 2 times than the combination of microbubble and ultrasonic ir- radiation group. While the recombinant gene ClpP2 could be expressed, and there was no significant difference between the two groups (Fig. 1). CONCLUSIONS Ultrasound microbubble could increase the efficiency of LFLIU mediated plasmid transfer to Mycobacterium tuberculosis. The results could provide experimental reference for LFLIU-mediated Fig. 1 (abstract P104). See text for description plasmid and DNA transformation into gram positive bacteria. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 117 of 122 P105 RESULTS After 2 weeks of focused ultrasound treatment, the rate of Novel ultrasound contrast agent based on lipid containing normal in treatment group was 75.00% (15/20) and that in control sinapultide microbubbles for potential theranostics group was 10.00%(2/20),there was statistically significant between two 1,2 1,2 1,2 Dong Liu , Fang Yang , Ning Gu groups (c2=17.289,P<0.05). The total number and the degranulated Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P105 number of mast cells, the expression of PAR2, SPand NK1-R in treat- ment group were significantly lower than those in control group (all School of Biological Science & Medical Engineering, Southeast P<0.05). University, Nanjing, China; State Key Laboratory of Bioelectronics and CONCLUSIONS Focused ultrasound can treat LSC through inhibiting Jiangsu Key Laboratory forBiomaterials and Devices, Nanjing, China PAR2, SP and NK1-R expression, decreasing the total number and the Correspondence: Dong Liu degranulated number of mast cells and reducing the inflammatory Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P105 reaction of tissues. OBJECTIVES Since the bubbles were reported to have the effects of contrast enhancement, gas-filled bubbles used as ultrasonic contrast P107 agents (UCAs) have been paid great attention [1, 2]. Apart from the Feasibility study of the induction of a neuronal response in L. usage of diagnostic agents, microbubbles have also widely used as Terrestris with ultrasound stimulation 1 1 1 drug and gene delivery vehicles [3]. Accordingly, the development of Jeremy Vion , William Apoutou N'Djin , Jean-Yves Chapelon , Jahan 2,3 novel UCAs have become one of the most clinical potential fields in Tavakkoli 1 2 ultrasound medicine. In this study, we developed a novel lipid con- U1032, Inserm, Lyon, France; Physics, Ryerson University, Toronto, taining sinapultide microbubbles filled with sulfur hexafluoride (SF6). Ontario, Canada; Institute for Biomedical Engineering, Science and The good stability as well as appropriately size indicating that they Technology, Toronto, Ontario, Canada have the potential to be used as ultrasound contrast agent in Correspondence: Jeremy Vion theranostics. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P107 METHODS Stable microbubbles were fabricated by adding sinapul- tide to lipid mixture and then sonicating to generate microbubbles OBJECTIVES Several works carried out these last years have demon- by cavitation from sulfur hexafluoride (SF6) in the mixture. strated the ability of ultrasound (US) to activate neurons at a systemic RESULTS The optimized preparation of sinapultide microbubbles had level. However, no complete description of the biophysical mechanisms mean diameter of 1.82 ± 0.15μm and zeta potential of −55.2 ± 3.9 involved in this phenomenon has been validated so far. In order to mV. The acoustic imaging analysis in vitro indicated that ultrasound identify experimentally the main mechanisms responsible for ultra- imaging enhancement could be acquired by both perfusion imaging sound neurostimulation, we here present an in-vivo study of the pos- and accumulation imaging (Fig. 1). sible effects of US exposures on the generation of Compound Action CONCLUSIONS In summary, a novel lipid microbubbles encapsulated Potentials (CAPs) in the common earthworm (Lumbricus Terrestris). synthesized pulmonary surfactant sinapultide were fabricated, which METHODS We chose to study in priority the effect of US radiation makes a promising clinical potential for future theranostics of ultra- force, as proposed by Wahab et al. (2012), on a simple giant nerve sound imaging and therapy. model: the ventral nerve cord of L. terrestris. The system used was a geometrically focused piezoelectric US transducer (radius of curva- ture: 46 mm, aperture: 50 mm) working at the frequency of 0.55MHz. In each trial, a medial section of the anesthetized worm was placed within the US focal area (geometry: ellipsoidal, axial length: 29 mm at -3dB, axial width: 1.9mm at -3dB), before being exposed to US se- quences similar to those proposed by Wright et al. (2015) on the per- ipheral crab nerve model (number of cycles: 25-28 cycles/pulse, PRF = 10 kHz, number of pulses: 80-200 pulses, peak-to-peak acoustic pressure: 4-14.4 MPa). The electrophysiological response of the nerve Fig. 1 (abstract P105). See text for description was monitored with two electrodes sunken through the animal in the vicinity of one end of the nerve. Before each trial, the functional- ity of the nerve was tested by applying anelectrical stimulation (pulsed, duration = 50 μs, amplitude = 5 V, PRF = 1 Hz, pulses num- P106 ber = 1 to 10) to the other end of the nerve. Complementarily, the Effects of focused ultrasound on expression of PAR2, SP and NK1-R response of the nerve to a mechanical stimulus applied to the super- in genital skin of SD rats with vulvar lichen simplex chronicus ior end of the worm was recorded. The electrical and mechanical Huan Yang, Huajun Tang, Yijin Fan, Chengzhi Li preliminary stimulations allowed repeatable observations of the CAP College of Biomedical Engineering, Chongqing Medical University, responses, which served as a reference for studying the nerve re- Chongqing, China sponse to US exposures. Correspondence: Huan Yang RESULTS The ultrasound sequences tested allowed inducing the gen- Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P106 eration of CAPs several times. The observed CAPs presented the same characteristics as those induced by the prior electrical and OBJECTIVES To explore the expression of protease activated receptor mechanical stimulations, as shown in Fig. 1. By computing the con- 2(PAR2), substance P(SP) and neurokinin-receptors in the genital skin duction velocity associated to the observed CAPs, we identifiedthe of SD rats with vulvar lichen simplex chronicus(LSC) after focused nervous structures stimulated to be the Medial Giant Fiber (MGF) ultrasound treatment. and the Lateral Giant Fiber (LGF). Over the trials, both fibers were METHODS Forty female rat models of LSC were established by stimulated at least once butnever simultaneously. However, a great Sally method, and then the rats were randomly divided into number of trials were performed, using various level of acoustic pres- treatment group (n=20) and control group(n=20). The rats in the sure, and this phenomenon appeared to be very littlerepeatable. treatment group were treated by focused ultrasound, and those CONCLUSIONS Several hypotheses were considered to explain the low in the control group received sham treatment. After focused repeatability of our US neurostimulation results. First, it may be neces- ultrasound intervention, the histological changes of vulva were sary to target specific areas in the giant nerve such as dorsal nodes, observed by using HE staining. The total number and the degra- which could be performed in our future trials by using ultrasound guid- nulated number of mast cells were identified by staining slides ance. Second, we cannot be sure that uncontrolled episodes of cavita- with toluidine blue. Immunohistochemistry was used to test the tion did not arise during some trials. This hypothesis is currently being expression of PAR2, SP and NK1-R. Differences among the groups tested in a series of trials where the ultrasound exposure is compatible were compared. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 118 of 122 with the controlled generation of cavitation. Finally, the myelin sheath RESULTS Ten specimens of PDA were scanned using HMI (5 from dis- embedding the giant nerve could also have an impact on the possibil- tal pancreatectomy and 5 from Whipple procedure). Preliminary hist- ity of inducing mechanical effects on the neural membrane. Hence, we ology results showed that the scan did not damage the tissue in any plan to adapt our set-up to the ventral nerve cord of the american lob- aspect, and did not impact the subsequent analysis. HMI was able to ster (Homarus Americanus). This unmyelinated nervous model has scan the full volume of the specimens with measurements asdeep as already been the object of preliminary studies performed during an 3 cm in the tissue. The whole range of stiffness from tumor to nor- international collaboration. In conclusion, we were able to demonstrate mal pancreas was assessed. The Fig. 1 shows the B-mode image of with this study the feasibility of inducing nervous responses with US the maximal cross-section of the tumor, and the HMI map overlaid and these responses were identical to those generated with electrical on the B-mode image. The mean displacement inside the tumor is and mechanical stimulation. Further investigations are necessary to 1.41 ± 1.00 μm versus 11.25 ± 3.69 μm in the surrounding tissue. control this phenomenon and make it more repeatable. This project Measurements in normal pancreatic tissue demonstrated a mean was supported by the Laboratory of Excellence (LabEx) DevWeCan and HMI displacement of 17.05 ± 3.98μm. A sharp mechanical contrast is the MitacsGlobalink Research Award - Campus France. evident on the HMI maps between the tumor and the surrounding tissue which correlates very well with the B-mode image. CONCLUSIONS This initial feasibility study showed that HMI is cap- able of scanning whole resected pancreatic cancer specimens with- out any damage to the tissue. The technique can assess deep parts of the tissue with greatly difference stiffness in a limited amount of time. HMI significantly detected PDA within the specimen with mean displacement lower by a ratio of 9.8 compared to the surrounding tissue and 12.1 compared to normal pancreas away from the mass. This indicates the possible presence of fibrotic tissue close to the tumor. The sharp tumor delineation observed on HMI maps was con- firmed on the B-mode image.HMI is thus shown promising for pan- creatic cancer detection and characterization. The technique can provide images of the entire organ and preliminary results indicate that it can distinguish PDA, fibrotic tissue from normal pancreatic tis- sue. HMI could thus constitute a very important tool for screening patients at risk of developing pancreatic cancer and showing the non-specific symptoms. Fig. 1 (abstract P107). Examples of nervous responses to an electrical stimulus (a), a mechanical stimulus (b) and an ulstrasound stimulus (c) P108 Fig. 1 (abstract P108). Pancreatic ductal adenocarcinoma in a distal Harmonic Motion Imaging (HMI) for pancreatic cancer detection pancreatectomy specimen. The Bmode image (top) is shown as well and characterization in post-surgical human specimens as the overlaid HMI displacement map (bottom) 1 2 1,3 Thomas Payen , Kenneth Olive , Elisa E. Konofagou Biomedical Engineering, Columbia University, New York, New York, USA; Medicine and Pathology & Cell Biology, Columbia University Medical Center, New York, New York, USA; Radiology, Columbia University Medical Center, New York, New York, USA Correspondence: Thomas Payen P109 Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P108 Experimental study on the influence of low-frequency and low- intensity ultrasound on the permeability of the mycobacterium OBJECTIVES Pancreatic ductal adenocarcinoma (PDA) has one of the smegmatis cytoderm and potentiation with levofloxacin 2 3 2 1 lowest prognosis due to non-specific symptoms leading to late diag- Yu Dong , Hang Su , Yonghong Du , Dairong Li nosis and therapeutic inefficiency. PDA is characterized by an un- Department of Respiratory disease, The First Affiliated Hospital of usually dense stroma limiting chemotherapy perfusion. Harmonic Chongqing Medical University, Chongqing, China; Chongqing medical Motion Imaging (HMI) assesses tissue mechanical properties by indu- university, Chongqing, China; Food and Drug Administration of Huiji, cing localized oscillation resulting from a periodic acoustic radiation Zhengzhou, China force. The amplitude of the induced displacement is directly related Correspondence: Yu Dong to the underlying tissue stiffness. The objective of this study was to Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P109 evaluate the utility of HMI in post-surgical human pancreatic cancer OBJECTIVES Tuberculosis (TB) is an infectious disease caused by the specimens and its capability of differentiating the tumor for perile- sional fibrotic tissue and normal tissue. bacterium Mycobacterium tuberculosis. The aim of this study is to in- vestigate the synergistic potentiating effect of low-frequency and METHODS A 4.5-MHz focused ultrasound transducer (FUS) generates an low-intensity ultrasound (LFLIUS) with levofloxacin on M. smegmatis amplitude-modulated beam resulting in harmonic tissue oscillations at its (MS) [a "surrogate of MTB"] and to explore underlying mechanisms. focus. Axial tissue displacement is estimated using 1D cross-correlation of RF signals acquired with a 2.5-MHz diagnostic transducer (P4-2, ATL) METHODS M. smegmatis was continuously irradiated by ultrasound using a plane-wave beam sequence, confocally aligned with the FUS.Ten of 42 kHz with several different doses (intensity and duration), human pancreatic specimens (approximate dimensions = 24 x 12 x 80 respectively. The bacteria vitality, structure and morphology were mm3) were assessed immediately after surgery. They were immersed in subjected to plate counting and microscopic examinations. Flow cy- PBS and their full volume was scanned for a maximum duration of 90 mi- tometry was adopted to test the fluorescence intensity of bacteria nutes. The imaging planes were chosen perpendicular to the pancreatic that were stained by FAD, or PI dye. Spatially and temporally aver- duct to correlate the HMI displacement withstandard histology analysis. aged intensity (ISATA) of 0.138 W/cm was then acted on Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 119 of 122 mycobacterium smegmatis combined with levofloxacin to confirm P111 the synergy between ultrasound and antibiotic. The safety on subsequent pregnancy in women after ultrasound RESULTS The results showed that the permeability of M. smegmatis ablation of uterine fibroids: a single-central retrospective study increased after ultrasound exposure; and the survival rate, structure Junshu Li, Wenzhi Chen, Jinyun Chen and morphology of bacteria of the low-dose (ISATA= 0.138 W/cm The State Key Laboratory of Ultrasound Engineering in Medicine Co- for 5 min) treated ultrasound group had no evident difference Founded by Chongqing and the Ministry of Science and Technology, compared with those of the control group (P>0.05) ; but the survival Chongqing Key Laboratory ofBiomedical Engineering, College of rate of bacteria significantly decreased and the bacteria structure got Biomedical Engineering, Chongqing Medical University, Chongqing serious damage in high-dose (ISATA= 0.329 W/cm for 20 min) Collaborative Innovation Center for Minimally-invasive andNoninvasive ultrasound group. The bactericidal effect of ultrasound combining Medicine, ChongQing, China with levofloxacin was evidently higher than that of single ultrasound Correspondence: Junshu Li or single levofloxacin. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P111 CONCLUSIONS A certain dose of LFLIUS can increase the permeabil- ity of M. smegmatis’ cytoderm and exert a synergy to bactericidal ac- OBJECTIVES A retrospective analysis to explore the impact of high- tion of levofloxacin. intensity focused ultrasound (HIFU) ablation of uterine fibroids in women on subsequent pregnancy. METHODS A retrospective analysis was conducted of women with uterine fibroids who underwent HIFU ablation at the Clinical Center for Tumor Therapy, Chongqing Medical University, Chongqing, China, P110 from January 1, 2010, to January 1, 2015. Inclusion criteria were: (1) Intracellular calcium response of hela cells stimulated by the aged 20-42 years (2) with fertility desire (3) have normal sexual life impulsive jetting flow from tandem bubble interaction without contraception after HIFU. Exclusion criteria were: fertility bar- 1 1 2 1 3 Fenfang Li , Chen Yang , Fang Yuan , Defei Liao , Guilak Farshid , Pei riers unrelated to HIFU, such as hysterectomy or bilateral oophorec- Zhong tomy. The registry of patient cases were screened by marital status Mechanical Engineering and Materials Science, Duke University, and birth history before treatment, and telephone follow-up informa- Durham, North Carolina, USA; HuaCells Corporation, Natick, tion including improvement of symptoms of uterine fibroids, sexual Massachusetts, USA; Department of Orthopaedic Surgery, Washington life, pregnancy, and delivery information after HIFU treatment. University, St. Louis, Missouri, USA RESULTS A total of 189 cases met inclusion criteria and followed-up ef- Correspondence: Fenfang Li fectively, the median follow-up time was three years. Majority of them Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P110 have symptoms of uterine fibroids improve significantly after HIFU treat- ment, and volume of uterine fibroids decrease(P<0.05). Among the 189 OBJECTIVES Cavitation plays an important role in the bioeffects pro- cases, there were 131 cases pregnancy with a total of 133 times.Of 131 duced by therapeutic ultrasound. However, limited studies have been pregnant women,19 were on-going pregnancy, 21 cases of abortion, the carried out to investigate the mechanisms of cellular mechanotransduc- remaining 91 cases in successful pregnancies and deliveries.The inci- tion that are activated by bubble pulsations in a controlled manner, espe- dence rates of complications during pregnancy and labor were 10.8% cially at the single cell level. Mechanotransduction is a process through (10/93) and 7.5% (7/93), respectively.The most common complications of which cells sense and respond to mechanical stimuli (e.g., membrane pregnancy in this study were placenta previa and placenta implant- tension and strain) and convert them to biochemical signals. In many ation(5/10).The incidence rate of massive hemorrhage during labor was cases, the initial mechanotransduction signal involves an increase in intra- 6.5%(6/93),the majority (66.7%) were associated with resection intramural cellular calcium ion (Ca2+) elicited by mechanical stress, which subse- fibroid during cesarean section.The study presented here did not contain quently triggers various downstream biochemical reactions that may any cases of uterine rupture during pregnancy or labor after HIFUtreat- alter the morphology and function of the cell. In this study, we examine ment. Ninety-one women successfully delivered 93 times with a cesarean the intracellular calcium response of single adherent HeLa cells evoked section rate of 72.0%(67/93). Among them,only fourteen women (20.9%) by the directional jetting flow from tandem bubbles (Rmax = 50 ± 2 μm). chose delivery by cesarean section for obstetric factors, while 79.1% METHODS Bubble dynamics, morphology, location, and adhesion chose delivery by cesarean section for social factors. conditions of individual cells were well-controlled in a microfluidic CONCLUSIONS In conclusion, HIFU is a safe and effective noninvasive system with surface patterning. Two types of calcium responses, in therapy to treat uterine fibroids in women who wish to retain the ability the form of intracellular calcium waves (ICW), were observed at dif- to conceive and deliver after treatment. Additionally, HIFU effectively pro- ferent standoff distance (Sd = 30 to 60 μm) of the tandem bubbleto vides remission of symptoms, has a low rate of complications through the cell through time-elapsed fluorescence imaging. pregnancy and labor, and does not increase the rate of spontaneous RESULTS The ICW was initiated from the leading edge facing the jet- abortion or delivery by cesarean section. Based on our findings presented ting flow, and propagated toward the trailing edge of the cell. At Sd/ here, we believe that HIFU should be recommended treatment for uter- Rmax = 0.6, a fast calcium response with short rise time and large amp- ine fibroids in women who wish to retain the ability to have children in litude in calcium concentration change was produced. In contrast, at the future, or who otherwise wish to not undergo hysterectomy. Sd/Rmax = 1, a slow calcium response with long risetime and small amplitude was often observed. In all cases, the calcium response was initiated by the influx of extracellular calcium through the membrane by either poration produced predominately at Sd/Rmax = 0.6 or ion P112 channel opening induced primarily at Sd/Rmax = 1.0, which were con- Low intensity pulsed ultrasound relieves myelosuppression firmed by mechanistic studies using calcium-free extracellular medium, induced by cyclophosphamide in rabbits membrane poration indicator propidium iodide, or the non-specific B. Liu mechanosensitive ion channel blocker Gd3+. The elicited calcium tran- Chongqing Medical University, Chongqing, China sient propagated in the cytosol through calcium-induced calcium re- Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P112 lease (CICR) from endoplasmic reticulum storage, which was inhibited by thapsigargin. We further demonstrated that attaching 6 μm integ- OBJECTIVES This study is to investigate the effect of low intensity rin-binding beads to the cell membrane can trigger calcium response pulsed ultrasound (LIPUS) on cyclophosphamide (CTX)-induced rabbit under conditions (Sd/Rmax = 1.2) that do not elicit such a response by myelosuppression. the tandem bubbles without the beads. METHODS A total of 40 New Zealand white rabbits were used to es- CONCLUSIONS These findings show that directional jetting flow from tablish the myelosuppression models by daily ear vein injection of 40 tandem bubbles can be used to induce highly controlled mechanical mg/kg CTX for 4 continuous days. They were randomly divided into stimulation on individual cells. LIPUS group and control group. LIPUS was performed once a day for Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 120 of 122 20 minutes each time for 7 and 28 days, respectively. The treatment METHODS Targestar-SA microbubbles conjugated with RGD li- effects of LIPUS were examined by comparing general conditions, gands wereusedto attach on thecell membraneofHeLa cells. mortality rates, blood routine and bone marrow hyperplasia in bone A Plasmid encoding for green fluorescent protein (GFP) was used marrow smears. The safety of the LIPUS irradiation was evaluated by for gene transfection. Two ultrasound conditions were selected to HE staining. stimulate two typical bubble activities while achieving good RESULTS LIPUS improved the number of peripheral blood cells and transfection efficiency and maintaining high cell viability, high bone marrow nucleated cells and thus reduced the mortality of pressure short pulse (0.6Mpa, 10μs) and low pressure long pulse model rabbits in different degrees. The irradiated parts of the skin (0.2MPa, 1ms). Multiply approaches were employed to investigate did not burn, and the muscle tissue in acoustic channels showed nor- bubble dynamics, cell membrane responses and plasmid DNA in- mal structures and no obvious pathological changes, suggesting ternalization routes: (i) Fast frame video microsocopy (Photron LIPUS irradiation is safe for the model rabbits. FASTCAM SA2) was used to record the microbubble dynamics. (ii) CONCLUSIONS LIPUS is a safe and effective method to relieve CTX- Real-time fluorescent microscopy of propidium iodide to indicate induced myelosuppression, which can be used for the prevention and and characterize pores. (iii) Confocal fluorescent microscopy of treatment of the occurrence and development of myelosuppression. plasmid DNA (Cy3 labled), cell membrane (Alexa Fluor 647 labled), and nuclear (Hoechst 33342 labeled) (Fig. 1). And (iv) SEM images for the surface topography of cell membrane. All P113 these observations were spatiotemporally correlated. Cost-effectiveness evaluated by markov decision-making tree RESULTS (i) When stimulated by a high pressure short pulse model of high-intensity focused ultrasound ablation versus (0.6Mpa, 10μs), targeted microbubble generates a transient and surgery for uterine fibroids reversible pore on the cell membrane, with radius varies from Liang Hu, Jinyun Chen, Yujie Feng, Chang Liu, Wenzhi Chen, Zhibiao several microns to several hundreds microns. While driven by low Wang pressure long pulse (0.2MPa, 1ms), the microbubbles undergo College of Biomedical Engineering, Chongqing Medical University, gentle oscillation, unlikely to perforate cell membrane. (ii) When Chongqing, China driven by high pressure short pulse, plasmid DNA can be directly Correspondence: Liang Hu and promptly propelled into the cytosol, even nuclear, through Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P113 pores with the diameter larger than about 100 microns. (iii) When driven by low pressure long pulse, the gentle bubble-cell inter- OBJECTIVES To investigate the cost-effectiveness of high-intensity fo- action significantly enhanced the plasmid DNA aggregation and cused ultrasound (HIFU) ablation versus surgery (myomectomy and retention onto the cell membrane, while the global endocytosis hysterctomy) for uterinefibroids. is responsible for plasmid DNA internalization. (iv) The determin- METHODS Direct medical cost, indirect cost, efficacy and safety out- ant factor for different plasmid DNA uptake processes is the comes of patients with uterine fibroids who ever underwent HIFU, ultrasound driven bubble- cell interaction. myomectomy, and hysterectomywere collected to build Markov Deci- CONCLUSIONS By establishing direct spatiotemporal correlation sion-making Tree Model based on health status of these patients. between ultrasound-stimulated targeted microbubbles activities The Model was employed to make simulation of health status transi- and the resulting intracellular plasmid DNA uptake process, this tion for uterine fibroids patients treated by these different proce- study revealed the complex and bubble-cell interaction dures. Cost-effectiveness, cost-utility as well as sensitivity analysis dependent nature of plasmid DNA uptake mechanisms at sin- were conducted for evaluation ofHIFU, myomectomy and hysterec- gle cell level. tomy for uterine fibroids. RESULTS The Markov Decision-making Tree Model simulation showed that each unit improvement of physical summary scale (PCS) and mental summary scale (MCS)related to HIFU ablation consumed 1782.16 and 1120.88 respectively. By contrast, each unit improve- ment of PCS and MCS related to surgery consumed 1825.43 and 1605.15 respectively. CONCLUSIONS The Markov Decision-making Tree Model simulation and healthy economic analysis demonstrate HIFU is dominatingly cost-effective for uterinefibroids. P114 Mechanisms of plasmid DNA intracellular uptake facilitated by ultrasound and targeted microbubbles is determined by bubble-cell interaction 1 2 2 1 N. Rong , H. Zhou , R. Liu , Z. FAN Biomedical Engineering, Tianjin University, Tianjin, Tianjin, China; College of Life Sciences, Nankai University, Tianjin, Tianjin, China Correspondence: N. Rong Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P114 OBJECTIVES Therapeutic Ultrasound combined with microbubbles can achieve non-viral, targeted delivery of genetic materials, as a promising delivery technique for gene therapy. However, insuffi- cient understanding of plasmid DNA intracellular uptake process Fig. 1 (abstract P114). Cy3 labeled plasmid DNA distribution in and the important role of microbubble plays during this process, Hela cells of different ultrasound parameter. Bright field images limits the progress of improving gene transfection efficiency and showing bubble information and representative confocal images translating this technique into clinics. Therefore, the aim of this showing pDNA distribution with Cy3 labeled pDNA (red), cell nuclei study is to investigate the intracellular uptake mechanisms of (blue) and membrane(yellow).(A) 0.6MPa,10μs, (B) 0.2MPa,1ms,(C) plasmid DNA mediated by targeted microbubbles driven by dif- Control without ultrasound stimulation.Scale bar is 5um ferent ultrasound conditions. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 121 of 122 P115 doses of 0 – 20 Gy. The analysis was conducted by using three differ- Optimization of tumour therapy by combination of focused ent cell lines for prostate cancer (PC-3, Vcap, LNcap), glioblastoma ultrasound and radiation guided by MRI and PET-MRI (LN405, U87, T98G) and head/neck tumor (FaDu, UT-SCC 5, UT-SCC 1 1 1 1 2 Doudou Xu , Lisa Landgraf , Xinrui Zhang , Michael Unger , Ina Patties , 8). Effects at the cellular level on metabolism (WST-1), proliferation 1 1 3,4 3,5 Johann Berger , Shaonan Hu , Lydia Koi , Antje Dietrich , Aswin (BrdU), membrane integrity (LDH release) and apoptosis (Annexin V) 3,4 3,4 1 1 Hoffmann , Mechthild Krause , Thomas Neumuth , Andreas Melzer were detected after treatment. Innovation Center Computer Assisted Surgery, Universität Leipzig, In vivo: PET-MR and MR guided focused ultrasound system allows Leipzig, Germany; Department of Radiation Therapy, Universität Leipzig, precise sonication treatment for small animals bearing tumors, under Leipzig, Germany; OncoRay - National Center for Radiation Research in real-time MR-thermometry [6]. Small animal PET-MR system (nanoS- Oncology, Dresden, Germany; Department of Radiation Oncology, can, Mediso) will be employed and integrated with a FUS transducer University Hospital Carl Gustav Carus, Technische Universität Dresden, (11×11 matrix array), which allows the function of beam forming to Dresden, Germany; German Cancer Consortium (DKTK), partner site achieve hyperthermia treatment. Local tumor irradiations under nor- Dresden, German Cancer Research Center (DKFZ), Heidelberg, Germany mal blood flow conditions will then be given with 200 kV X-rays (0.5 Correspondence: Doudou Xu mm copper -filter) and 20 mA at a dose rate of ~ 1.1 Gy/min (X-ray Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P115 machine type Yxlon Y.TU 320-D03) [7]. The self-filtering of the X-ray tube is 3 mm beryllium and 3 mm aluminum. Nude mice bearing Introduction heterotopic tumors on hind leg will be treated by both FUS and RT The two ZIK-Centers for Innovation Competence, ICCAS in Leipzig in sequence. Tumor size monitoring, histology study and UPLC mo- and OncoRay in Dresden, have joined forces to start a new multidis- lecular analysis will be investigated post the combination therapy. ciplinary 6.3 million Euro research project: SONO-RAY - Tumor ther- Robot installed in PET-MR: A MR-compatible robotic arm system TM apy combining image-guided (PET-MR and MR) focused ultrasound (INNOMOTION , Innomedic) [8, 9] was installed with Biograph mMR and radiation therapy. The goal of SONO-RAY is to combine noninva- MR-PET (Siemens Healthineers) in the Department of Nuclear Medi- sive image-guided therapy approaches of magnetic resonance cine of the University Medical Center Leipzig to investigate the ef- guided focused ultrasound and radiation therapy to improve the effi- fects of a combination of focused ultrasound and radiation therapy. cacy of cancer treatment. The robotic arm will reposition the ultrasound transducer during the sonication treatment. It is possible to detect residual tumor tissue Hypothesis and Motivations: after the treatment by using PET-MR imaging to provide an optimal SONO-RAY project aims to combine therapeutic focused ultrasound treatment outcome. (FUS) and radiation therapy (RT) to treat malignant solid tumors and MR guided FUS-Sonalleve: MR-guided high-intensity focused ultra- tumor metastases. The hypothesis underlying this approach is that sound (MR-HIFU) is a noninvasive technique for depositing thermal the combination of two tissue-destroying energies (the energy of energy in a controlled manner deep within the body. The Philips high-intensity acoustic waves and ionizing radiation) is more effect- Sonalleve MR-HIFU system, initially released in 2010, was installed in ive in cancer treatment than the effect of employing one of the Leipzig University Hospital at the beginning of 2017, introduced a above two energy forms alone [1-3]. The central scope of the project new approach for uterine fibroids[10, 11] and bone metastasis treat- is to develop tumor cell biology fundamentals, the computer-aided ment [12, 13] by employing thermal ablation. Sonalleve system also model formation and to evaluate the success potential of a future offers solutions of hyperthermia research platform [14, 15] in combin- clinical use of a FUS-RT combination therapy. Within the framework ation with radiation therapy and chemotherapy in cancer treatment. of this project, the basic principle of the combined FUS and RT effect MR guided FUS-prostate system: The TULSA-PRO System (Profound on tumor cells (in vitro) and the tumor tissue (in vivo) is going to be Medical) is a transurethral MR guided FUS system for whole gland investigated preclinically. Clinical applications could be the treatment ablation of the prostate [16]. A test system was installed at the uni- of various tumors of the head and neck, prostate carcinoma and versity hospital Dresden to perform the world’s first compatibility glioblastoma. tests on a Philips Ingenuity TF PET/MR scanner. The system com- Project work package: prises a transurethral ultrasound applicator with 10 FUS elements In order to provide the basis for the combined FUS-RT method, in vitro working at 4 or 14 MHz, depending on the penetration depth re- and in vivo experiments are being be carried out to elaborate the ther- quired to heat the rim of the prostate (i.e. the so-called “thermal ab- mal and mechanical effect on the tumor tissue [4]. After identification lation boundary”) up to 55 deg Celcius. The system has an of the thermal and mechanical individual parameters for FUS, these are endorectal cooling device to prevent the rectal mucosa from over- transferred into simulation models in order to predict the behavior of heating. The ultrasound applicator also has a cooling circuit to spare the fabric on FUS[5]. The simulation models are validated using the pa- the urethra. The power and frequency of the 10 ultrasound transduc- rameters. Based on the data obtained above, FUS and RT will be gener- ers are individually steered by real-time MR-thermometry. ated spatially dispersed biologically active doses. Algorithms will be used for the anatomical co-registration and accumulation of the bio- logically active dose distributions generated by FUS and RT on MR im- References ages. Furthermore, software modules will be developed and validated, [1] R. Cirincione, F.M. Di Maggio, G.I. Forte, L. Minafra, V. Bravata, L. Castiglia, which support the clinical user in therapy planning. Optimization of ap- V. Cavalieri, G. Borasi, G. Russo, D. Lio, C. Messa, M.C. Gilardi, F.P. plication sequences of the FUS-RT technology and the incorporation in Cammarata, High-Intensity Focused Ultrasound- and Radiation Therapy- therapy into the treatment process (clinical workflow) of the patient will Induced Immuno-Modulation: Comparison and Potential Opportunities, also be investigated and developed. Ultrasound in medicine & biology, 43 (2017) 398-411. Experimental facilities in SonoRay [2] T. Refaat, S. Sachdev, V. Sathiaseelan, I. Helenowski, S. Abdelmoneim, M.C. In vitro FUS and RT: A high throughput in vitro 96-well sonicator Pierce, G. Woloschak, W. Small, Jr., B. Mittal, K.D. Kiel, Hyperthermia and was designed and allows individual sonication for each of the wells radiation therapy for locally advanced or recurrent breast cancer, Breast, in a 96-well plate. It consists of 96 single transducers at an operating 24 (2015) 418-425. frequency of 1 MHz and a maximum energy of 0.05 W/cm . A 150 kV [3] S.K.Wu, C.F.Chiang,Y.H. Hsu, H.C. Liou, W.M. Fu,W.L.Lin, X-ray machine (DARPAC 150-MC) was employed for irradiation at Pulsed-wave low-dose ultrasound hyperthermia selectively enhances Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 122 of 122 nanodrug delivery and improves antitumor efficacy for brain P116 metastasis of breast cancer, Ultrasonics sonochemistry, 36 Combination therapy for cancer by PET-MR and MR image guided (2017) 198-205. focused ultrasound and radiation 1 1 1 1 2 [4] J. Beik, Z. Abed, F.S. Ghoreishi, S. Hosseini-Nami, S. Mehrzadi, A. Shakeri- Doudou Xu , Lisa Landgraf , Xinrui Zhang , Michael Unger , Ina Patties , 1 1 3,4 5 Zadeh, S.K. Kamrava, Nanotechnology in hyperthermia cancer therapy: Johann Berger , Shaonan Hu , Lydia Koi , Marc Fournelle , Steffen 5 1 1 From fundamental principles to advanced applications, Journal of con- Tretbar , Thomas Neumuth , Andreas Melzer trolled release : official journal of the Controlled Release Society, 235 Innovation Center Computer Assisted Surgery, Universität Leipzig, (2016) 205-221. Leipzig, Germany; Department of Radiation Therapy, Universität [5] J. Hu, Y. Ding, S. Qian, X. Tang, Simulations of adaptive temperature Leipzig, Leipzig, Germany; OncoRay - National Center for Radiation control with self-focused hyperthermia system for tumor treatment, Ul- Research in Oncology, Dresden, Germany; Department of Radiation trasonics, 53 (2013) 171-177. Oncology, University Hospital Carl Gustav Carus, Technische Universität [6] A.J. Loeve, J. Al-Issawi, F. Fernandez-Gutierrez, T. Lango, J. Strehlow, S. Dresden, Dresden, Germany; Fraunhofer IBMT, St. Ingbert, Germany Haase, M. Matzko, A. Napoli, A. Melzer, J. Dankelman, Workflow and inter- Correspondence: Doudou Xu vention times of MR-guided focused ultrasound - Predicting the impact Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P116 of new techniques, Journal of biomedical informatics, 60 (2016) 38-48. [7] A. Yaromina, T. Kroeber, A. Meinzer, S. Boeke, H. Thames, M. Baumann, D. INTRODUCTION FUS-RT stands for the combination therapy of fo- Zips, Exploratory study of the prognostic value of microenvironmental cused ultrasound (FUS) and radiation therapy (RT). The aim of parameters during fractionated irradiation in human squamous cell SONORAY project is to use the synergistic effect of heat and carcinoma xenografts, International journal of radiation oncology, mechanical effects of sound as well as ionizing radiation in order biology, physics, 80 (2011) 1205-1213. to improve the results of radiooncological treatments of malig- [8] A.J. Krafft, J.W. Jenne, F. Maier, R.J. Stafford, P.E. Huber, W. Semmler, M. nant solid tumors and metastases. Physical and biological effects Bock, A long arm for ultrasound: a combined robotic focused ultrasound are being investigated and the synergy of FUS and RT will be setup for magnetic resonance-guided focused ultrasound surgery, Med- quantified in simulation, cell, and small animal studies. The work ical physics, 37 (2010) 2380-2393. is accompanied by the development of multimodal treatment [9] N.V. Tsekos, A. Khanicheh, E. Christoforou, C. Mavroidis, Magnetic planning and information systems to prepare for a seamless inte- resonance-compatible robotic and mechatronics systems for image- gration into the clinical workflow. The project aims at developing guided interventions and rehabilitation: a review study, Annual review of a proof-of-concept system and workflow for the translation into biomedical engineering, 9 (2007) 351-387. clinical use. [10] Y.S. Kim, B. Keserci, A. Partanen, H. Rhim, H.K. Lim, M.J. Park, M.O. Kohler, METHODS A high throughput in vitro sonicator with 1.14 MHz single Volumetric MR-HIFU ablation of uterine fibroids: role of treatment cell transducer made by piezoelectric ceramic material was employed size in the improvement of energy efficiency, European journal of radi- and allows individual sonication for wells in a 96-well plate. T98G gli- ology, 81 (2012) 3652-3659. oma cells were exposure to acoustic intensity of 71 W/cm with [11]M.J. Voogt, H. Trillaud, Y.S. Kim, W.P. Mali, J. Barkhausen, hyperthermia (40-45°C) duration of 134 sec, and 109 W/cm with L.W. Bartels, R. Deckers, N. Frulio, H. Rhim, H.K. Lim, T. Eckey, hyperthermia (40-45°C) duration of 65 sec, respectively. A 150 kV X- H.J. Nieminen, C. Mougenot, B. Keserci, J. Soini, T. Vaara, M.O. ray machine (DARPAC 150-MC) was employed for irradiation at doses Kohler, S. Sokka, M.A. van den Bosch, Volumetric feedback of 0-20 Gy to investigate the radiation curve of T98G cells. For FUS ablation of uterine fibroids using magnetic resonance-guided and RT combinations, 5 and 10 Gy were used to treat T98G cells high intensity focused ultrasound therapy, European radiology, 24hrs post sonication. Effects at the cellular level on metabolism 22 (2012) 411-417. (WST-1), proliferation (BrdU) and membrane integrity (LDH release) [12]M. Huisman, M.K. Lam, L.W. Bartels, R.J. Nijenhuis, C.T. Moonen, were detected after treatment. F.M. Knuttel, H.M. Verkooijen, M. van Vulpen, M.A. van den Bosch, RESULTS The preliminary RT results showed dose dependent loss in Feasibility of volumetric MRI-guided high intensity focused ultrasound cellular NAD(P)H levels of 60% for T98G cell lines at 20 Gy. A release (MR-HIFU) for painful bone metastases, Journal of therapeutic of LDH was observed from 4% (0 Gy) to 17% (20 Gy). The highest im- ultrasound, 2 (2014) 16. pact of RT was detected during analysis of DNA synthesis (BrdU) [13]M.K. Lam, M. Huisman, R.J. Nijenhuis, M.A. van den Bosch, which nearly stopped at dosages above 5 Gy. In terms of the first M.A. Viergever, C.T. Moonen, L.W. Bartels, Quality of MR thermometry FUS and RT combination experiment, there is no higher LDH release during palliative MR-guided high-intensity focused ultrasound in the combination therapy group in comparison to FUS or RT single (MR-HIFU) treatment of bone metastases, Journal of therapeutic treatment groups. In contrast, cells treated by combination of FUS ultrasound, 3 (2015) 5. hyperthermia at 40-45°C for 65 secs with 10 Gy irradiation treatment [14]E.J. Dorenberg, F. Courivaud, E. Ring, K. Hald, J.A. Jakobsen, E. suggested lower cell viability of 81% (WST-1 assay) in comparison to Fosse, P.K. Hol, Volumetric ablation of uterine fibroids using treatment of FUS hyperthermia only (97% viable cells) and RT only Sonalleve high-intensity focused ultrasound in a 3 Tesla scanner– (90% viable cells). first clinical assessment, Minimally invasive therapy & allied tech- CONCLUSIONS In conclusion, longer FUS hyperthermia duration nologies : MITAT : official journal of the Society for Minimally Inva- from 100 sec to 1000 sec should be further investigated in sive Therapy, 22 (2013) 73-79. combination with RT treatment. The interval between FUS [15] N.M. Hijnen, E. Heijman, M.O. Kohler, M. Ylihautala, G.J. Ehnholm, A.W. hyperthermia and irradiation treatment will be conducted within Simonetti, H. Grull, Tumour hyperthermia and ablation in rats using a 8hrs according to literatures. A fitted treatment for different cell clinical MR-HIFU system equipped with a dedicated small animal set-up, lines will be necessary based on the different radiosensitivities. International journal of hyperthermia : the official journal of European So- Future in vitro investigations of effects of FUS hyperthermia as ciety for Hyperthermic Oncology, North American Hyperthermia Group, well as of a combined therapy on other tumor cells need to be 28 (2012) 141-155. conducted. [16] M. Burtnyk, T. Hill, H. Cadieux-Pitre, I. Welch, Magnetic resonance image guided transurethral ultrasound prostate ablation: a preclinical safety and Publisher’s Note feasibility study with 28-day followup, The Journal of urology, 193 (2015) Springer Nature remains neutral with regard to jurisdictional claims in published 1669-1675. maps and institutional affiliations. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Therapeutic Ultrasound Springer Journals

Abstracts from the International Society for Therapeutic Ultrasound Conference 2017

Journal of Therapeutic Ultrasound , Volume 6 (1) – May 15, 2018

Loading next page...
 
/lp/springer-journals/abstracts-from-the-international-society-for-therapeutic-ultrasound-WElcBxGJPk

References (36)

Publisher
Springer Journals
Copyright
Copyright © 2018 by The Author(s).
Subject
Medicine & Public Health; Ultrasound; Imaging / Radiology
eISSN
2050-5736
DOI
10.1186/s40349-018-0110-x
Publisher site
See Article on Publisher Site

Abstract

Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 https://doi.org/10.1186/s40349-018-0110-x MEETING ABSTRACTS Open Access Abstracts from the International Society for Therapeutic Ultrasound Conference 2017 Nanjing, China. 31 May - 02 June 2017 Published: 21 May 2018 RESULTS 18F-DPA-714 uptake was increased at the sonication Oral Presentations sites in all locations (Fig. 1). The ratio of the percent increase in SUV between pFUS+MB treated striatum and hippocampus to O1 contralateral side is depicted in graph (mean+/-SEM) clearly Neuroinflammation after disrupting the blood brain barrier with showing large increase in uptake for both regions compared to pulsed focused ultrasound and microbubbles imaged by 18F-DPA- normal brain. The neuroinflammatory changes persisted for at 714 PET and MRI least 14 days after 2 weekly sonications. The coefficient of vari- Zsofia I. Kovacs, Georgios Z. Papadakis, Tsang-Wei Tu, Sanhita Sinharay, ation for PET scans was <10%. This corresponded to Iba1 activa- William C. Reid, Bobbi Lewis, Dima A. Hammoud, Joseph A. Frank tion visible on histology. Figure 2 contains normalized National Institutes of Health, Bethesda, Maryland, United States histograms from VOI for Group 2 and Group 3 rats derived from Correspondence: Zsofia I. Kovacs pFUS+MB treated (ipsilateral) and contralateral brain that shows Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O1 a shift to lower T2* values for sonicated regions. OBJECTIVES Blood brain barrier (BBB) disruption with MR-guided pulsed focused ultrasound (pFUS) and microbubbles (MB) has CONCLUSIONS Rats receiving pFUS+MB to open the BBB showed a been advocated as a noninvasive adjuvant treatment for malig- clear upregulation of TSPO expression consistent with microglial nancies and neurodegenerative diseases. A sterile inflammatory activation/neuroinflammation, even after one sonication session. reaction has been recently described in the brain as a result of Histograms derived from T2* maps MRI clearly shows that sonication pFUS+MB (Kovacs et al. 2016). However, one potential issue of with BBB algorithm results in left shift in T2* values that would be weekly pFUS+MB treatments is the lack of data on the long- consistent with hypointense voxels on T2*w MRI and abnormatilies term effects on inflammation. The purpose of this study was to on histolology. These preliminary results contradict current assump- evaluate the effects of multiple weekly courses of pFUS+MB tions that the effects of pFUS+MB are confined primarily to the endo- exposures in the rat brain using micro-Positron Emmision Tom- thelium and vessel wall. Further assessment of the long-term effects ography (PET) and 18F-DPA-714, a marker of translocator pro- of pFUS+MB is necessary before this approach can be widely imple- tein (TSPO) upregulation/microglial activation and an indication mented in clinical trials. of neuroinflammation. METHODS Female rats were assigned to three different groups References based on the number of weekly pFUS+MB: Group 1: pFUS+MB Kovacs, Z. I., et al. (2017). "Disrupting the blood-brain barrier by focused ultrasound x1, PET scans performed 24 hours later (n=6) ; Group 2: pFUS induces sterile inflammation." Proc Natl Acad Sci U S A 114(1): E75-E84. +MB x2 with PET scans performed within 10 days after 2nd son- O'Reilly, M. A. and K. Hynynen (2012). "Blood-brain barrier: real-time feedback- ication (n=5) ; and Group 3: pFUS+MB x6 with PET scans per- controlled focused ultrasound disruption by using an acoustic emissions- formed 7-9 days later (n=5). The left striatum (str) and right based controller." Radiology 263(1): 96-106. hippocampus (hc) were targeted in all animals. 100 μlof MB (OptisonTM, GE Healthcare, Little Chalfont, UK) was administered intravenously over 1 minute starting 30 secs before pFUS. Acous- tic energy was delivered to the brain using “BBB configuration function” based on algorithm reported (O’Reilly et al. 2012) to de- and termine optimal acoustic pressure for BBB opening via 1.5f 2.5f ultra harmonic acoustic emission detection for every single pulse (9 focal points, 120 sec/9 focal points – striatum, 120 sec/4 focal points – hippocampus) using an 825 kHz hydrophone with a single-element spherical FUS transducer (center frequency: 589.636 kHz; focal number: 0.8; aperture: 7.5 cm; RK-100, FUS Instruments, To- ronto, Ontario, Canada). T2* map were created from multiecho gradient echo sequence at 3T (Achieva, Philips Healthcare, Andover, MA) through the rat brain with TE=7 msec, echo train length 5 and echo spacing 7 and Tr=1500 msec. T2* maps were created by fitting signal intensity at each voxel to a single exponential fit with in-house software Fig. 1 (abstract O1). 18FDPA714 PET scans (coregistered to MRI and histogram analysis was performed on volume of interests (VOI). template) in two rats: A: Hippocampus uptake 24 hours after one Static microPET/CT scans emission data was acquired 30-60 min after sonication. B: Striatum uptake 10 days after six weekly sonications. C: injection of 18F-DPA-714. VOIs were drawn in the targeted areas and Graph of % difference between sonicated ipsilateral and contralateral uptake was compared to the contralateral unaffected side. Uptake striatum and hippocampus regions for Groups 1, 2 and 3 values were normalized to cerebellum. © The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 2 of 122 fiber structure- and gray matter-abnormalities on DTI MRI were de- tected in regions with the T2* abnormalities suggestive of increased astrogliosis and transient axonal damage. MRI findings following pFUS+MB x6 demonstrated more pronounced evidence of damage within the parenchyma and atrophy. FDG-PET did not show differ- ences between sonicated and contralateral cortex or hippocampus at any time point. BrdU showed evidence of increased neurogenesis in pFUS+MB treated regions. CONCLUSIONS The long term effects of pFUS+MB exposures in rats revealed that both single and repeated pFUS+MB cause structural in- jury at the location of sonication up to 13 weeks post treatment based on advanced imaging techniques. Histological evidence Fig. 2 (abstract O1). Mean normalized histograms derived from VOI showed that associated with the pathological changes observed by from pFUS+MB treated cortex/striatum and hippocampus (ipsilateral) MRI, there was evidence of neuronal damage and loss, neurogenesis compared to contralateral brain forGroup 2 and 3 cohorts of rats. and activated microglia. These results suggest the importance of There is clear shift to lower T2* values for sonicated regions for 2x v long term monitor of the brain following low intensity pFUS+MB be- fore its clinical translation. O2 Long term effects of pulsed focused ultrasound and microbubbles References detected by multivariate imaging modalities Arvanitis, C. D., et al. (2016). "Cavitation-enhanced nonthermal ablation in Zsofia I. Kovacs, Tsang-Wei Tu, Georgios Z. Papadakis, William C. Reid, deep brain targets: feasibility in a large animal model." J Neurosurg Dima A. Hammoud, Joseph A. Frank 124(5): 1450-1459. National Institutes of Health, Bethesda, Maryland, United States Downs, M. E., et al. (2015). "Long-Term Safety of Repeated Blood-Brain Barrier Correspondence: Zsofia I. Kovacs Opening via Focused Ultrasound with Microbubbles in Non-Human Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O2 Primates Performing a Cognitive Task." PLoS One 10(5): e0125911. OBJECTIVES Blood brain barrier (BBB) opening by Guided Pulsed Fo- cused Ultrasound (pFUS) and microbubbles (MB) is a non-invasive O3 treatment of various central nervous system diseases. However, the Characterization of different microbubbles in assisting focused potential adverse effects of repeated pFUS+MB exposure have not ultrasound-induced blood-brain barrier opening 1 2, 3 3, 4 5 been thoroughly elucidated and may limit clinical translation. To date Sheng-Kai Wu , Po-Chun Chu , Wen Yen Chai , Shih-Tsung Kang , 3 5 5 3 MRI scans of repeated BBB opening by pFUS+MB have been Chih-Hung Tsai , Ching-Hsiang Fan , Chih-Kuang Yeh , Hao-Li Liu achieved without hemorrhage, edema and behavioral changes in Institute of Biomedical Engineering, National Taiwan University, Taipei, non-human primates (Arvanitis, et al. 2016; Downs, et al. 2015). By in- Taiwan; Department of Research and Development, NaviFUS corp, corporating detailed multimodal imaging, we characterized the long Taipei, Taiwan; Department of Electrical Engineering, Chang-Gung term effects of single or repeated pFUS+MB in the rat brain. The pur- University, Taoyuan City, Taiwan; Department of Diagnostic Radiology pose of the study is to reveal the morphological and pathological and Intervention, Chang-Gung Memorial Hospital, Taoyuan City, Taiwan; changes following repeated BBB opening in the striatum and hippo- Department of Biomedical Engineering and Environmental Sciences, campus as monitored by 3T and 9.4T MRI, FDG-positron emission National Tsing Hua University, Hsinchu, Taiwan tomography (PET) and histology over 13 weeks. Correspondence: Sheng-Kai Wu TM METHODS pFUS+MB (Optison , GE Healthcare, Little Chalfont, UK) Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O3 0.3 – 0.5 MPa, 10 ms burst length, 1 % duty cycle, 9 focal points, 120 sec/9 focal points – striatum, 120 sec/4 focal points – hippocampus, This abstract is not included as it has already been published: 589.636 kHz; focal number: 0.8; aperture: 7.5 cm; (FUS Instruments, Wu S-K, Chu P-C, Chai WY, Kang S-T, Tsai C-H, Fan C-H, Yeh C-K, Liu H-L. Toronto, Ontario, Canada) was targeted in female rats (n=6/group) Characterization of Different Microbubbles in Assisting Focused either once or six weekly to the striatum and the contralateral Ultrasound-Induced Blood-Brain Barrier Opening. Sci Rep. 2017; 7. Available hippocampus. 100 μL of MB were administered intravenously over 1 from: http://www.nature.com/articles/srep46689 doi:10.1038/srep46689 minute starting 30 secs before pFUS. Rats received 3 daily doses of 300mg/kg 5-Bromo-2′-deoxy-uridine (BrdU, Sigma-Aldrich, St. Louis, O4 MO) intraperitoneally before sonication to label proliferating cells Development of an A-Synuclein (SNCA)-based mouse model for in vivo. T2, T2* and Gd-enhanced T1-weighted images were obtained Parkinson's disease by ultrasound-guided CNS delivery by 3.0T MRI (Achieva, Philips Healthcare, Andover, MA), T2, T2* and 1 2 2 Chung-Yin Lin , Yu-Chien Lin , Hao-Li Liu diffusion tensor imaging (DTI) were performed by 9.4T MRI (Bruker, Institute for Radiological Research, Chang Gung University, Taoyuan Billerica, MA). Parameters for DTI: 3D spin echo EPI; TR/TE 700 ms/37 2 2 City, Taiwan; Department of Electrical Engineering, Chang Gung msec; b-value 800 sec/mm with 17 encoding directions; voxel size University, Taoyuan City, Taiwan 200 μm (isotropic). Fractional anisotropy (FA) and the asymmetry of Correspondence: Chung-Yin Lin magnetization transfer ratio (MTRasym) were derived for mapping Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O4 structural injury and glucose levels. Rats received ~1.1 mCi of 18F- FDG via intravenously to quantitate glucose uptake by PET/CT OBJECTIVES Parkinson’s disease (PD) is the second most common (Inveon, Siemens, Munich, Germany). PET emission data was acquired neurodegenerative disease characterized by the progressive degen- for 60 min. Dynamic images were reconstructed and image analyses eration of dopaminergic neurons in the substantia nigra (SN) and the was performed (PMOD Technologies Ltd., Zurich, Switzerland). presence of α-synuclein-containing inclusion bodies in the cytoplasm Animals were euthanized 7 or 13 weeks after the first pFUS of neurons. In this study, we propose to develop a novel asynuclein treatment. Histological evaluation of brain and tracking of BrdU over-expression PD mouse model via ultrasound-guided CNS delivery tagged cells was performed. of SNCA gene. RESULTS Gd-enhanced T1-weighted images after each sonication METHODS A plasmid encoding both the green fluorescent protein demonstrated BBB disruption in the striatum and the hippocampus (GFP) gene and the SNCA gene was prepared. A gene-liposome sys- at 3T. Gd T1 enhancement, T2 and T2* abnormalities were not seen tem, in which the liposomes are designed to carry SNCA plasmid in the brain 1-day post pFUS+MB at 9.4T MRI. Hypointense regions DNA, forms liposomal-plasmid DNA (LpDNA) complex. Ultrasound appeared on T2* MRI 2 weeks post pFUS+MB consistent with devel- (US) exposure used the SonoPore KTAC-4000 to induce BBB opening opment of microhemorrhages within the parenchyma. White matter Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 3 of 122 (1.0-MHz, voltage = 85 V, burst length = 10 ms, 10% duty cycle, and optimization procedure, from the experimentally determined Dox 3-min sonication duration). The longitudinal expression of GFP was kinetics under the different conditions tested. Finally, based on quantitated via an in vivo imaging system (IVIS). The SNCA gene ex- the fitted values, a sensitivity analysis was performed and the pression level was confirmed via immunoblotting, and histological most important parameters that affect the Dox transport and staining will be used to identify transfected cells via fluorescent mi- cellular uptake in the tumor interstitium were determined. croscopy. Dopamine and metabolic DOPAC protein levels in the RESULTS Multiphoton microscopy revealed up to one order of mag- brain were determined via HPLC. nitude higher Dox extravasation after FUS- BBB/BTB disruption as RESULTS With longitudinal observation of IVIS monitoring, animals compared to control. In addition, a five-fold increase of Dox penetra- with US treatment showed significant promotion of LpDNA release tion was found after FUS- BBB/BTB disruption compared to control into the SN and demonstrated enhanced expression of genes upon (>100μm vs. <20μm, based on Dox penetration regression). The nu- sonication with US-BBB opening, while both the GFP and α-synuclein merical model indicated that only the vessel diffusion coefficient 2 2 protein expression were successfully measured via Western blotting. (4.71 ± 1.7μm /s Vs 0.41 ± 0.1μm /s, a tenfold increase) and the Immunoblotting and histological staining confirmed the expression pressure difference (4.3 ± 0.45mmHg vs 2.9 ± 0.32mmHg) were of reporter genes in neuronal cells. significantly different between the FUS- BBB/BTB disruption and the CONCLUSIONS This study suggests that IV administration of LpDNA control. While the increase in vessel diffusion coefficient was in combination with US-BBB opening can provide effective gene de- anticipated, the increase in pressure difference, which lead to a livery and expression in the SN, demonstrating the potential to system’s Peclet number greater than one (Pe>1), suggested a achieve non-invasive and targeted gene delivery for α-synuclein fundamental change in the interstitial transport mechanism. over-expression PD mouse model. Assessment of the temporal evolution of the drug concentration in the interstitial space in the experimental data verified that the transport after FUS-BBB/BTB disruption was convection dominated. O5 Single cell analysis revealed significant Dox uptake by endothelial cells Experimental and numerical assesment of Doxorubicin (EC), suggesting that microbubble vibrations can lead to significant pharmacokinetics in brain metastasis from breast cancer after changes in cellular transmembrane transport. Interestingly, in the FUS focused ultrasound-induced blood-brain/blood-tumor barrier treated animals’ stroma cells (SC) appeared to take-up the drug at disruption higher rate than the endothelial cells (uptake slope 0.44 vs 0.93 in EC 1,4 2 1 2 Costas Arvanitis , Vasileios Askoxylakis , Yutong Guo , Jonas Kloepper , and SC respectively). These differences in the uptake curves led to 2 3 2 5 Meenal Datta , Miguel Bernabeu , Dai Fukumura , Nathan McDannold , significant changes in the rate of cellular transmembrane transport in Rakesh Jain the numerical model during parameter fit. Finally, sensitivity analysis Mechanical Engineering, Georgia Institute of Technology, Atlanta, showed that FUS-BBB/BTB disruption makes drug uptake more Georgia, USA; Radiation Oncology, Massachusetts General Hospital, sensitive to all the parameters of the system, suggesting that the main Harvard Medical School, Boston, Massachusetts, USA; Centre for Medical barrier to drug transport has been overcome. Interestingly, the BBB/BTB Informatics, University of Edinburgh, Edinburgh, UK; Biomedical permeability remains an important parameter of the system even after Engineering, Georgia Institute of Technology, Atlanta, Georgia, USA; FUS, indicating that further disruption may be beneficial. Radiology, Brigham and Women's Hospital, Harvard Medical School, CONCLUSIONS The in vivo and in silico data demonstrate signifi- Boston, Massachusetts, USA cant changes in the tumor microenvironment after FUS-BBB/BTB Correspondence: Costas Arvanitis disruption. The most notable changes included: i) increase in Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O5 BBB/BTB permeability, ii) transition to convection dominated drug transport, and iii) increased cellular transmembrane transport in OBJECTIVES Blood brain and blood tumor barriers (BBB and BTB) endothelial cells and stroma cells. Sensitivity analysis showed that constitute a major obstacle to the transport of therapeutics in brain the system has become more amenable to interventions, suggest- tumors. Focused ultrasound (FUS), when combined with circulating ing that FUS can lead to the development of new therapeutic microbubbles, provides a noninvasive method to locally and transi- strategies to treat brain tumors. ently disrupt the BBB/BTB. While several studies have demonstrated its potential for targeted drug delivery, there is a lack of fundamental understanding of the impact of this method on the pharmacokinetics O6 of anticancer agents in the brain microenvironment. In this study, we Acoustic emissions during blood-brain barrier disruption with examine the impact of FUS-induced BBB/BTB disruption on the trans- focused ultrasound and real-time feedback control under infusion port of the chemotherapeutic agent doxorubicin (Dox) in a model of administration of microbubbles – feasibility study in rodent model 1 1 1 1 breast cancer brain metastases using intravital microscopy and drug Chenchen Bing , Debra Szczepanski , Imalka Munaweera , Yu Hong , 2 1,2 transport mathematical modeling. Ian Corbin , Rajiv Chopra METHODS Human HER2-amplified and estrogen dependent BT474 Radiology, UT Southwestern Medical Center, Dallas, Texas, USA; breast cancer cells, genetically modified to express green fluorescent pro- Advanced Imaging Research Center, UT Southwestern Medical Center, tein, were stereotactically implanted in the brain of mice with cranial win- Dallas, Texas, USA dows.Atatumorsizeof~20-40mm , BBB/BTB disruption was performed Correspondence: Chenchen Bing using FUS exposures by a custom built portable FUS system and Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O6 concurrent i.v. administration of microbubbles. To cover the entire tumor and its periphery four non-overlapping sonications were OBJECTIVES Glioblastoma multiforme (GBM) is the most lethal pri- performed. Shortly after sonication, the auto-fluorescent mary brain tumor, with aggressive and fatal progression of disease chemotherapeutic agent Dox (7.5mg/kg) was administered i.v. inevitable. During the earlyinfiltration of GBM cells into normal brain The pharmacokinetics of Dox, including intratumoral uptake and regions surrounding the primary tumor, this peritumoral environ- clearance, were measured for 15 minutes using intravital ment has an intact blood-brain barrier (BBB) which inhibitsthe deliv- multiphoton microscopy. The Dox cellular kinetics were also ery of therapeutic agents. BBB disruption with focused ultrasound determined. For mathematical analysis, we developed a numerical could be used to achieve peritumoral delivery of therapeutic agents model combining the Navier-Stokes and Brinkman equations for flow to tackle infiltrative cancerprogression. However, variability in vascu- modeling in the tumor vasculature and interstitial space, coupled with lature structure and function around a tumor leads to heterogeneous a convection-diffusion-reaction model of drug transport based on perfusion of microbubbles which can impact the acousticthresholds Michaelis-Menten kinetics. The model parameters (vessel permeability, required for BBB disruption. A more reliable means of BBB disruption interstitium porosity, rate of cellular transmembrane transport, tissue is desired. Prior studies suggest that by integrating an acoustic emis- hydraulic conductivity, and pressure difference driving flow across the sion detector and afeedback control algorithm to evaluate the inten- vessel wall and interstitium) were inferred, using a numerical sity of acoustic emission at sub-/ultra-harmonics, more controlled Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 4 of 122 and consistent BBB opening can be achieved. The goalof this study O7 was to evaluate the acoustic emissions and BBB opening in rodents Unilateral focused ultrasound-induced blood-brain barrier opening and validate the feasibility of real-time feedback control under intra- redistributes hyperphosphorylated Tau in an Alzheimer's mouse venous infusionadministration of Optison microbubbles. model 1 1 1 3 METHODS A custom-built focused ultrasound transducer with a central Maria Eleni Karakatsani , Tara Kugelman , Shutao Wang , Karen Duff , 1,2 frequency of f = 0.5MHz was attached to a stereotactic system and Elisa E. Konofagou used to deliver ultrasoundenergy into the target brain region. One Biomedical Engineering, Columbia University, New York, New York, USA; 2 3 piezocomposite hydrophone with resonant frequency at 0.75MHz was Radiology, Columbia University, New York, New York, USA; Pathology built to acquire the signals emitted from stimulatedmicrobubbles. A and Cell Biology, Columbia University, New York, New York, USA feedback control algorithm was implemented in LabVIEW to quantify Correspondence: Maria Eleni Karakatsani the area under curve (AUC) within sub-/ultra-harmonic bands during Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O7 theultrasound exposure and to adjust the focal pressure accordingly based on the difference between current AUC and a desired threshold. OBJECTIVES Focused ultrasound has been shown to interact with Initial in vitro tests wereperformed in which microbubble was infused Alzheimer’s pathology and particularly to trigger a mechanism that into a single tube (0.08 mL/min, tubing I.D. = 1mm), and the AUC was results in the reduction of the amyloid plaque load. However, a less quantified at 0.5f ,1.5f and 2.5f as a function offocal pressure and studied interaction is that of ultrasound with the tangle formation 0 0 0 microbubble concentration. Due to the significant response detected at that has been implicated in the cognitive decline of Alzheimer’s pa- 1.5f (0.75MHz), this specific frequency band was used for controlling in tients. Tau pathology can be characterized by increased density of thefollowing experiment. The ability to maintain acoustic emissions at a the hyperphosphorylated tau protein that results in tangle formation. target AUC level was also evaluate in vitro. Next an in-vivo study was At the early stages of Alzheimer’s disease, tau protein can be local- performed in a rat model toevaluate the acoustic emissions and the ized primarily in the axons while in late pathology, somatodendritic feasibility of real-time control of the AUC at a target level. Acoustic tau is more pronounced. With the current study we investigate the emissions from a bolus injection and continuous infusionwere evalu- interaction of focused ultrasound-induced blood-brain barrier open- ated. For bolus injection, a fixed pressure level of 0.54 MPa was applied, ing with the tau distribution. Moreover, the unilateral sonication of while for infusion experiment, the feedback control was used to control the transgenic brain provides a unique opportunity to explore poten- the AUC atvarious levels. Evans blue dye was used as an indicator of tial bilateral effects. BBB opening. METHODS For this study the initial cohort included 10 mice of the RESULTS A minimum transmit focal pressure of 0.33 - 0.41 MPa was rTg4510 line (3.5 months old), 5 of which were randomly assigned to required to trigger the bubble activity in the sub/ultra-harmonic the control group that did not receive any sonication and 5 to the bands in vitro, with the greatestchanges occurring at 0.75 MHz (1.5f ) treatment group. The treatment group received a double sonication (Fig. 1B). With varying concentration of microbubbles, the feedback covering almost the entire hippocampal region once per week for 4 control algorithm could adjust the focal pressure accordinglyto consecutive weeks. The day after the last sonication the mice were achieve a consistent harmonic emission (Fig. 1C). By observing the sacrificed. The brains were sectioned and counterstained for tau pro- in-vivo AUC response of bolus injection used previously to success- tein (AT8) as well as microglia activation (CD68). The images were ac- fully open the BBB, a thresholdAUC value was selected at 1.5f and quired by means of confocal microscopy over a z-series to account tested with a continuous infusion of microbubbles. Successful main- for depth differences and enabling co-visualization of the tau protein tenance of the AUC at different target value was achieved invivo at and microglia distribution. A custom algorithm was constructed to multiple locations in the brain, and BBB opening was confirmed as quantify the number of cells and the axonal distribution of the tau- leakage of Evans Blue at the target locations (Fig. 1D). marker. Background noise was automatically removed by color-based segmentation using k-means clustering and the cells were detected CONCLUSIONS In this study, a stereotactic BBB opening system was by the Hough transform. The axons were quantified based on their extended to incorporate a feedback control algorithm based on the length marked by the tau protein. The same brain slices were utilized sub-/ultra-harmonic emissionsfrom microbubbles. The AUC responses to quantify the hippocampal density of the CD68 marker by inten- were characterized and stable feedback control was demonstrated sity-based quantification. both in-vitro and in-vivo. Using infusion injection instead ofbolus in- RESULTS Figure 1 shows two representative examples of the con- jection, combined with the real-time feedback control system, a trol and the treatment group. Following the hippocampal forma- more reliable BBB opening can be achieved across a larger brain re- tion of the control brain, it can be observed that both gion for applications focusedon drug delivery to peritumoral regions. somatodendritic and axonal tau (red) are evident. In particular, the tau marker engulfs the cell bodies and the entire in-plane length of the axons can be detected. Although the cell bodies af- fected by tau protein are also evident in the animals that re- ceived four sonications, the axonal tau was less pronounced. More specifically, the axonal distribution of the tau protein was not continuous. Differences across hemispheres were only de- tected in the treatment group. Moreover, the phagocytic micro- glia (green) seem almost absent in the control brains while they can be observed in both hemispheres of the treatment group. Quantification of the tau distribution and density are currently ongoing. CONCLUSIONS Focused ultrasound-induced blood-brain barrier opening affects tau distribution after unilateral application. Axonal tau was significantly less pronounced in the sonicated region. Whether the tau protein reduced after sonication has been trans- ferred to other cell bodies or has been entirely eliminated from the brain remains to be determined by the ongoing quantification process. Microglia were clearly activated by the sonication but its re- Fig. 1 (abstract O6). See text for description lationship to tau re-distribution remains to be established. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 5 of 122 of the discrepancies indicates that there is not any systematic error associated (mean spatialerror tends toward zero). Simulations in test cases also validated the potential of the method by showing its ac- curacy in a wide range of configurations. CONCLUSIONS A method of cavitation localization based on a three hydrophone network was explored. Localizations performed with this method were corroborated by high-speed observation in water. Also, simulations show that the method can be applied in a wide range of cases. It was shown that the use of a relatively narrow frequency bandwidth around the sub-harmonic frequency permitted to en- Fig. 1 (abstract O7). Hippocampal formation counterstained for tau hance the accuracy of the method. It is hypothesized that this comes protein with AT8 (red) and microglia activation with CD68 (green). from the fact that this signalis emitted only by oscillating bubbles The first row corresponds to the ipsilateral and contralateral side of and not by reflections of the excitation signal. In addition to the re- a control brain and their magnified regions. The second row sults shown in this study, this method could be augmented by a corresponds to the ipsilateral and contralateral side of a treated complete characterization of the cavitation event by the analysis of brain and the corresponding magnified regions the cavitation signals. Also, by adding a hydrophone, 3-dimensional localization can be performed at minimal numerical cost. The ob- tained results and the possibilities that could be explored in the fu- ture give credit to this technique as a versatile tool for O8 cavitationposition monitoring, a major step for clinical transfer in all Evaluation of a three hydrophone-based method for cavitation applications related to cavitation. localization 1 2 2 1 Maxime Lafond , Cyril Lafon , Jean-Louis Mestas , Shin-ichiro Umemura Graduate School of Biomedical Engineering - Umemura-Yoshizawa Laboratory, Tohoku University, Sendai, Miyagi, Japan; LabTAU, INSERM U1032, Université deLyon, Lyon, France Correspondence: Maxime Lafond Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O8 OBJECTIVES In the context of clinical applications, the complete characterization of the cavitation activity includes also the cavitation localization. However, the current methods for cavitation localization present various disadvantages such as the limited bandwidth of the echo imaging probes, high numerical cost and the need forexpensive equipment. This limits strongly the transfer of cavitation-based appli- cations toward their clinical use. We propose to apply a source localization algorithm to cavitation monitoring and characterization. In the context of cavitation-related therapy, it is expected to increase the reliability level cavitation applications, giving it morecredit for clinical transfer. In order to overcome these limitations, we intend to use a PVDF hydrophone network to localize the bubble cloud, con- sidered as a simple acoustic source. A classic method for source localization is triangulation. The localization of the cavitation cloud is deduced from the delays ob- tained between threereceptors with known positions. In our case, Fig. 1 (abstract O8). Comparison between highspeed camera the receptors are PVDF hydrophones. Two confocal transducers are observation and the position of the cavitation cloud calculated emitting a pulse at 1.1 MHz in order to generate cavitation in the op- with the method based on three hydrophones. While the algorithm tical field of a high-speed camera. The signals from the three hydro- based on the full frequency bandwidth (red) gives an approximate phones were recorded during the US pulse on a digital oscilloscope position of the cavitation cloud, the one based on a bandwidth and the delays between the hydrophones were calculated by finding reduced around the subharmonic frequency (blue) give an enhanced the delay maximizing the inter-correlation between the recorded sig- localization of the cavitation cloud. The green circle denotes for the nals. The source position calculated from thedelays was finally super- focal point of the US apparatus, used for the superimposition of the imposed over the images from the camera (Fig. 1). The positions calculated position over the highspeed images calculated with this method were compared to the positions of the clouds visually estimated. The mean discrepancy was calculated. The method was firstly applied using the signals with full frequency bandwidth. Then, the post-processing operation was repeated after keeping only the bandwidth of 200 kHz around the sub-harmonic frequency (550 kHz). Also, simulations are performed to evaluate the O9 versatility of the method in various test cases. Notably, spatial Design study and experimental validation of a novel phased array spreading of the source, source separation and the influence of the based on Fermat’s spiral hydrophones repartition are evaluated. 1 1 2 1 Pascal Ramaekers , Martijn de Greef , Rémi Berriet , Chrit Moonen , RESULTS The high-speed camera observations were compared with Mario Ries the localization technique for each one of 8 independent pulses. The 1 2 UMC Utrecht, Utrecht, Netherlands; Imasonic SAS, Voray-sur-l'Ognon, position of the cavitation cloudis calculated with a discrepancy of 3.1 France ±1.8 mm in the case of the full frequency bandwidth. By processing Correspondence: Pascal Ramaekers the data only in the 200 kHz frequency band around the 550kHz Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O9 sub-harmonic, the accuracy is improved to 1.4±0.8 mm. The analysis Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 6 of 122 OBJECTIVES Previous work has shown numerically that a phased extracorporeal HIFU applications the beam steering capabilities of array transducer based on Fermat’s spiral could potentially pro- the tessellated arrays are sufficient within the relevant steering vide improvements for extracorporeal HIFU therapy, in particular range. Out of the three tessellated designs, the array based on φ with respect to the energy exposure of the near field. In a first = 137.51° is technically the most favorable as it has the most step, this design study numerically evaluated phased array de- uniform element size distribution. The tapered array performs signs based onFermat’s spiral for three different spirals and two slightly better in terms of beam steering, but also induces higher different element types. The transducers were evaluated on three pressure levels in the near field. different aspects: element size distribution, beamsteering capabil- ities, and pressure distribution in the near field. In a second step, one of the designs was evaluated experimentally through hydro- phone measurements. METHODS For the numerical evaluation of the different phased array designs, the arrays were populated by 256 elements and evaluated for two different element types: circular elements of 3.8 mm diameter (sparse arrays), and element shapes generated by applying Voronoi tessellation to the predefined element po- sitions (tessellatedarrays). Three different spirals were used to determine the phased array element positions. The first spiral had a divergence angle of φ = 87.85°. Such a design has been- shown to minimize peak grating lobe levels in a study on 2D ultrasound imaging arrays based on Fermat’sspiral. Thesecond spiral was constructed using φ = 137.51° (the golden angle). Previous work has shown that using the golden angle optimizes the packing efficiency of the spiral, which ensures the most uni- form element size distribution. This has practical implications with respect to construction and electrical matching of the Vor- onoi tessellated array. The third spiral was also based on the- golden angle, but with a Taylor tapering window characterized by a sidelobe level of -30 dB applied. It has been shown for array antennas that such a tapering windowreduces sidelobe levels of the array. All transducers were designed using an aper- ture diameter and radius of curvature of 16 cm (f-number = 1). An overview of theevaluated tessellated arrays is shown in Fig. 1. Acoustic simulations were performed under free-field condi- tions using propagation of the angular spectrum of planewaves. For the experimental validation of one of the designs, pressure measurements were conducted as a function of instantaneous Fig. 1 (abstract O9). See text for description acoustic output power using a fiberoptic probe hydrophone (FOPH 2000, RP Acoustics). RESULTS Considering the element size distribution for the Voronoi tessellated arrays, the array based on φ = 137.51° was shown to be the most uniform, with all element sizes within 20% difference from the mean element size. The array based on φ = 87.85° had 6 notable outliers, but was otherwise comparable to the array basedon φ = 137.51°. The array based on φ = 137.51° and a tapering window was shown to have a highly nonuniform distribution of element sizes, with only 89 elements within 20% difference from the mean element size. Maximum pressure levels obtained for off-axis sonications are shown in Fig. 2. The sparse arrays performed comparable in terms of beam steering, with the tapered array giving a slightly better per- formance than the other two. For the tessellated arrays the tapered array gave the best off-axis performance, followed by the array base- don φ = 137.51°. The worst performance is observed for the array based on φ = 87.85°. Figure 3 shows cumulative histograms of pres- sure levels in the near field (10 < Z < 90 mm prefocal) for an on axis sonication for each of the six arrays. The sparse arrays gave a highly similar performance in terms of near field pressure distribution. The tessellated arrays substantially reduced near field pressure levels compared to thesparse arrays. Comparing the tessellated arrays, the best performance was observed for the array based on φ = 137.51. Figure 4 shows pressure as a function of time for the hydrophone measurements on one of the tessellated transducers for different in- stantaneous acoustic output powers. Rarefactional and compressional pressures ranged from 14.47 and 66.53 MPa respectively at 100 W, to 40.00 and 237.50 MPa at 800 W. CONCLUSIONS Tessellated arrays provide an improvement over sparse arrays in terms of near field energy exposure. This comes Fig. 2 (abstract O9). See text for description at the cost of reduced beam steering capabilities, but for Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 7 of 122 OBJECTIVES Even if bubble collapses are commonly thought to be the key element of cell permeabilization for drug delivery applica- tions, recent works have demonstrated the possibility of transfecting cells by gentle oscillating bubbles (stable cavitation), possibly result- ing in lower cell lysis or tissue damages. Nevertheless, in a bubble cloud, both stable and inertial cavitation activities would naturally co- exist, thus making difficult to quantify the contribution of both re- gimes on the drug deliveryprocess. In this work, we present a regulation system aiming at distinguishing the two regimes and con- trolling the stable cavitation activity during time. METHODS A feedback-loop process is implemented on the level of the subharmonic component emitted from a bubble cloud and recorded by a needle hydrophone (Fig. 1). The cloud is created by a focused transducer emitting a 550 kHz pulsed sine wave (duty cycle: 0.2; cycle duration: 250 ms; sonication duration: 60 s) in a watertank, either in the bulk (free configuration) or in a plastic tube located at the focal point (vessel-confined configuration). The feedback-loop performs a real-time modulation of the applied voltage at a 250μs loop rate, allowing to control the subharmonic emission (SC index), as well as measuring iner- tial cavitation activity (broadband noise emission, IC index). RESULTS Without regulation (open loop), the measure of the mean SC is not reproducible for a given acoustic intensity (Fig. 2-a), and the SC and IC indices oscillate during all the sonication time (Fig. 2- b). In contrast, the regulation system leads to reproducible results, showing a SC index which always reaches the feedback-looptarget value (Fig. 2-c) and remains constant during time (24 dB in Fig. 2-d). Fig. 3 (abstract O9). See text for description Moreover, as showed in Fig. 2-c, the feedback-loop allows achieving a high subharmonic response (up to 20dB) without broadband noise emission (0-5 dB), thus limiting the inertial cavitation effect. Finally, for a given high SC level, the results show a gain up to 20% of acous- tic energy in comparison to the sonication without regulation. CONCLUSIONS Evidences of control of the stable cavitation activity are reported, associated with (1) the control of the stochastic behavior of the bubble population, (2) a stable SC index during time, (3) the minimization of inertial cavitation activity, thus limiting the destructive effects due to bubble collapses, and (4) the saving of acoustic energy required for initiating stable cavitation. [Work supported by the French National Research Agency, LabEx CeLyA (ANR-10-LABX-0060) and granted bythe ANR-MOST project CARIBBBOU (ANR-15-CE19-0003)] Fig. 4 (abstract O9). See text for description O10 Real-time control of the stable cavitation activity in free and vessel-confined bubble clouds 1 1 1 Corentin Cornu , Matthieu Guedra , Jean-Christophe Bera , 2 3 1 Wen-Shiang Chen , Hao-Li Liu , Claude Inserra Univ Lyon, Université Lyon 1, INSERM, LabTAU, F-69003, LYON, France, Lyon, France; Department of Physical Medicine & Rehabilitation, National TaiwanUniversity Hospital, Taipei, Taiwan; Department of Fig. 1 (abstract O10). Schematic representation of the experimental Electrical Engineering, Chang Gung University, Taoyuan City, Taiwan invitro device with focused transducer in a water tank. Realtime Correspondence: Corentin Cornu monitoring and feedbackloop process are described Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O10 Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 8 of 122 to predetermined targets to achieve multi-position neurostimulation. According to the classical ultrasound imaging technique by array transducer, special array transducer can be used to perform dual functions of ultrasound imaging and neurostimulation. This study in- vestigates the feasibility of using linear array ultrasound transducer system to achieve image-guided dual-target ultrasound brain stimu- lation on mouse. METHODS A dual modes ultrasound system, showed as Fig. 1(A), is developed to perform B-mode imaging for predetermination of stim- ulated target and to produceacoustic radiation force for neurostimu- lation. The stimulation position was controlled by beamformer inside the field programmable gate array device. Brain imaging was done first and followed by image guided stimulation. 5 male BALB/c mice, 8-10 weeks old, 20g (+/-25%) in weight were used. Mouse was anes- thetized byintraperitoneal injection of ketamine (70mg/kg) and xyla- zine (7mg/kg) cocktail. The mouse's hair was cropped by scissor and removed by depilatory. The mouse was laid prone on an automatic Fig. 2 (abstract O10). Reproducibility experiments in pulsed sonication heating pad (69002, RWD Life Science Co.) at 37 degrees centigrade for the SC (subharmonic emission level) and IC (broadband noise level) and with its head gently immobilized using stereotaxic frame (68028, indicators as a function of the applied voltage in open loop (2-a) and the RWDLife Science Co.). Different positions were selected to evaluate SC target value in closed loop (2-c). Examples of temporal stability curves the effect. Stimulated effect was evaluated by Electromyography sig- for the evolution of the SC and IC indicators versus time at the applied nals recorded by an electrophysiological acquisition system (MedLab- voltage 180 mV in open loop (2-b) and at the SC target value 24 dB in U8C502, MedEase Ltd., Nanjing, China). The motion responses of closed loop (2-d) mouse were captured by a camera (HD1080P, AoniLtd., Shenzhen, China) in real-time. RESULTS Figure 1(B) demonstrates a B-mode image of a mouse brain for predetermining two different stimulation targets. The exact posi- tions of the targets can be selected by the computer mouse, and are pointed out by the arrows. Figure 1(C) indicates the different effects of the EMG responses evoked by the ultrasound stimuli on the target O11 A and B. Our results indicate that imaged-guided dual-target brain Research platform for rodent studies of wavefront engineered stimulation by array ultrasound can evoke different motion responses ultrasonic neuromodulation on mouse. 1 1,3 1 1 1 Steve Krupa , Eilon Hazan , Omer Naor , Michael Plaksin , Inbar Brosh , CONCLUSIONS The imaged-guided dual-target brain stimulation sys- 2 2 1 3 1 Noam Maimon , Yoav Levy , Eitan Kimmel , Itamar Kahn , Shy Shoham tem can achieve dual targets brain stimulation with varying spatial Department of Biomedical Engineering, Technion I.I.T, Haifa, Israel; and temporal conditions underthe guidance of B-mode imaging, 2 3 Insightec LTD, Tirat HaCarmel, Israel; Department of Medicine, which offers a useful tool for the applications on neural circuits stud- Technion I.I.T., Haifa, Israel ies and clinical therapies. Correspondence: Steve Krupa Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O11 This abstract is not included as it has already been published: Shoham S, Krupa S, Hazan E, Naor O, Levy Y, Maimon N, Brosh I, Kim- mel E, Kahn I. A126 Research platform for rodent studies of wave- front engineered ultrasonic neuromodulation. J Ther Ultrasound. 2016; 4(Suppl 1):31. Available from: https://jtultrasound.biomedcen- tral.com/articles/10.1186/s40349-016-0076-5 O12 Image-guided dual-target brain stimulation on mouse by array ultrasound Guofeng Li, Jiehan Hong, Qiuju Jiang, Peitian Mu, Ge Yang, Congzhi Wang, Weibao Qiu, Hairong Zheng Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences, Shenzhen, China Correspondence: Guofeng Li Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O12 Fig. 1 (abstract O12). (A) shows the block diagram of the proposed imagedguided dualtarget ultrasound brain stimulation system. Figure OBJECTIVES Ultrasonic neuromodulation has become a promising (B) demonstrates a Bmode image of a mousebrain for predetermining approach for neural science and clinical application. Currently single two different stimulation targets. The exact positions of the targets can element focused ultrasound transducer was widely used for perform- be selected by the computer mouse, and are pointed out by the ing ultrasonic stimulation. However, the stimulated position is fixed, arrows. Figure(C) indicates the different effects of the EMG responses and is hard to be precisely identified, when performing the brain evoked by the ultrasound stimuli on the target A and B stimulation. It is important to precisely steer the focus of ultrasound Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 9 of 122 O13 capabilities of the system were preserved. Among the tested gates, 2ms Pseudo-random gated sonications for reducing cavitation during period pseudo-random codes seem to be the modulation that increases thermal ablation in the brain the cavitation threshold the most while preserving heating capabilities. 2,1 3 3 3,4 Cyril Lafon , David Moore , Matthew Eames , John Snell , James Indeed, very fast switches may be associated to electronic difficulties for 2 3,4 Larner , Neal Kassell non-demonstrated benefit in terms of cavitation reduction. Work spon- 1 2 LabTAU, INSERM, Lyon, France; Department of Radiation Oncology, sored by the FUS Foundation through the Robert Merkin Fellowship and University of Virginia, Charlottesville, Virginia, USA; FUS Foundation, performed with the technical support of InSightec. Charlottesville, Virginia, USA; Department of Neurosurgery, University of Virginia, Charlottesville, Virginia, USA Correspondence: Cyril Lafon O14 Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O13 A flat array with 4096 elements for MR-guided focused ultrasound therapy 1 1 1 OBJECTIVES Transcranial HIFU is now used in clinics for treating es- Yuexi Huang , Ben Lucht, Rohan Ramdoyal , Samuel Gunaseelan , 1 1 1,2 sential tremor and proposed for many other brain disorders. This Tyler Portelli , Ping Wu , Kullervo Hynynen 1 2 promising treatment modality still faces several limitations. HIFU-in- Sunnybrook Research Institute, Toronto, Ontario, Canada; Department duced thermal ablation in the brain requires high energy resulting of Medical Biophysics, University of Toronto, Toronto, Ontario, Canada eventually in undesired cavitation and potential side effects. Original Correspondence: Ben Lucht strategies should be tested to increase treatment safety and efficacy. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O14 The goals of the present work were: 1- to evaluate the potential in- crease of the cavitation threshold using pseudo-random gated soni- OBJECTIVES MR-guided focused ultrasound has been demonstrated cations and 2- to assess the heating and steering capabilities with in various applications for non-invasive thermal ablations. The such sonications. The performance of pseudo-random sonications current clinical devices for body applications use spherically curved was compared with conventional continuous exposures. phased arrays, which improve focus quality for a limited number of METHODS The experiments were performed with the transcranial transducer elements (~200), at the expense of a limited steering MR-compatible ExAblate Neuro system (InSightec). It is a 1024-elem- range, therefore mainly rely on mechanical positioning for covering a ent, 30 cm diameter, 15 cm focal length, transducer operating at a large treatment volume. In this study, we developed a fully popu- frequency of 660 kHz. Four methods of sonication were compared: lated flat array, which allows for amuch wider steering range. Feasi- continuous wave (CW), gated emissions with pseudo-random 2ms (p- bility using this new design for thermal ablation and hyperthermia Rand2ms) or 33 us (p-Rand33us) -period codes and 30 kHz square over large target volumes was demonstrated in animal studies. (Squ30kHz). The duty cycle (DC) was set to 50% for the gated sonica- METHODS A flat array of 14 cm in diameter with 4096 elements was tions. The cavitation threshold was first evaluated in water. For each manufactured in house with center-to-center element spacing of condition, electrical driving power was increased step by step from half-wavelength at the centre frequency of approximately 500 kHz. 20 to 500W and the acoustical noise was recorded with the inte- Custom driving electronics were built using Application-Specific Inte- grated passive cavitation detectors. The spectral energy was aver- grated Circuit (ASIC) technology. The MR-compatible arrayand driving aged over the 250-500 kHz bandwidth and the sonication time (10s). system were mounted on the standard bed of an MR scanner Heating trials were then performed in a hydrogel tissue mimicking (MR750, GE Healthcare, Milwaukee, WI, USA). In vivo studies were material (TMM, ATS Laboratories). The electrical power was set to 15, performed on 16 pigs on thigh muscles. Acoustic power up to 240W 30 or 60 W, the exposureduration to 9 or 18 s. For a fair comparison over 50s were applied with MR thermometry monitoring. Focus was with CW, 50% DC sonications had either the electrical power or the either stationary during sonications or steered in circularpattern lat- exposure duration doubled. The temperature was measured by MR- erally or along the acoustic beam. The minimum delay between thermometry when focusing at the geometrical focus and when steering spots for loading new phasing parameters was 3 ms. Lesions steering the beam off-focus by 5mm-steps. were measured by T2-weighted imaging (FRFSE, TR 5000ms, TE RESULTS The assumption was made that the occurrence of cavitation 100ms). For hyperthermia applications, multi-foci sonication pattern resulted in an inflexion of the scattered energy vs. power curve. Due was applied to achieve heating over a large volume (30cm3) for 15 to frequency modulation, harmonics occur in the bandwidth of inter- min at 43°C. est for the p-Rand33us and Squ 30kHz sonications and their energy RESULTS The array focused energy well by using geometric beam- increased also with power. The threshold was then harder to assess. steering phase delays and was able to provide sustained acoustic Sudden changes in the scattered energy occurred at electrical pow- power with repeated treatments.Temperature was above 60°C at focus. ers of 180, 400, 500, 400W for CW, p-R and 2ms, p-Rand33us and Good volumes of thermal coagulation were achieved with a wide steer- Squ30kHz respectively. Unsurprisingly, higher thresholds were mea- ing range as large as 12 cm. For hyperthermia, heating around 43°C sured when repeating the experiments in more degassed water. over the large volume was observed by MR thermometry. These settings did not allow to heat the TMM when electronically CONCLUSIONS In this study we demonstrate for the first time that fully steering the focus more than 15mm-off the geometrical focus. Doub- electronically steered arrays are feasible. With a flat design and transducer ling the exposure duration did not allow to compensate the 50% DC element spacing at half-wavelength, wide steering was achieved without because of thermal diffusion. At a power of 30W, a maximal sacrificing focus quality. A wide steering range also eliminates the need temperature rise of 9°C was measured with gated sonications applied for a motorized positioning system, which simplifies system design and al- for18s while CW 30W sonication for 9 s gave a 13°C increase in lows precise MR thermometry without motion induced artifacts. temperature. However, the heating patterns obtained at 60W with p- R and 2ms and 30 W in CW were very similar for a constant 9s dur- ation. For the same conditions, lower temperature rises were mea- O15 sured for p-R and 33us and Squ30kHz. Several assumptions can be Effects of rarefactional pressure from pulsed focused ultrasound in made to explain this difference: the gating is too rapid for reaching murine melanoma and breast tumor models: implications for steady state, the amplifier struggles to switch so quickly, or cavitation immunotherapy enhanced heating is reduced. Omer Aydin, Scott R. Burks, Saejeong Kim, Joseph A. Frank CONCLUSIONS Coded sonications were proposed using rapid ran- Radiology and Imaging Sciences, NIH Clinical Center, Bethesda, dom phase shifts (Chapelon et al. 1996) or chirp (Tang and Clement Maryland, USA 2009). In the present work, it has been demonstrated on a clinical Correspondence: Omer Aydin Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O15 device that randomly gated signals also increase the cavitation threshold in water. The general idea consists in breaking the dynamic OBJECTIVES Focused ultrasound (FUS) has long been used for ther- of the bubbles moving from monochromatic to broadband sonica- mal ablation of malignant tumors. The potential role for nonthermal tions. It was then necessary to check that the heating and steering Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 10 of 122 pulsed FUS (pFUS) as amethod to modulate the innate immune re- O16 sponse to tumors is under investigation. We have previously charac- Boiling histotripsy ablation of renal carcinoma in the Eker rat terized the molecular changes in numerous tissues and have produces significant changes in the immune system 1 1 2 3 demonstrated changes in the tissue microenvironment and immune George R. Schade , Wayne Brisbane , Stella Whang , Yak-Nam Wang , 2 4 5 3 cell phenotypes following pFUS. However, the parameters needed to Kayla Gravelle , Venu Pillarisetty , W. Conrad Liles , Vera Khokhlova , 3 2 2 maximize pFUS-inducedmolecular changes in tumor microenviron- Michael Bailey , Tatiana D. Khokhlova , Joo Ha Hwang ments that could modulate immunotherapeutic response are essen- Department of Urology, University of Washington, Seattle, Washington, tially uninvestigated. We treated the B16 murine melanomaand 4T1 USA; Department of Gastroenterology, University of Washington, breast cancer flank tumors with pFUS of 1 MHz over a range of peak Seattle, Washington, USA; Center for Industrial and Medical Ultrasound, negative pressures (PNP) (1–8 MPa) without microbubble infusions University of Washington, Seattle, Washington, USA; Department of and evaluated the molecular response to sonication. Surgery, University of Washington, Seattle, Washington, USA; METHODS Subcutaneous B16 or 4T1 tumors were established in the Department of Medicine, University of Washington, Seattle, legs of C57BL/6 (B16) or BALB/c (4T1) mice and allowed to reach ~8 Washington, USA mm in diameter. Magnetic resonance-image-guided pFUS with passive Correspondence: George R. Schade cavitation detection (RK100, FUS Instruments, Toronto, ON, Canada) or Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O16 ultrasound-image-guided pFUS (VIFU, Alpinion Medical Systems, Both- ell, WA) were administered at 1 MHz over a range of PNP values (1, 2, OBJECTIVES Focused ultrasound (FUS) exposures have demonstrated 4, 6, and 8 MPa). One hundred sonications were delivered to each ras- the ability to modulate the immune system to stimulate an anti-tumor terpoint using a 10% duty cycle and a pulse repetition frequency of 10 immune response. Ourgroup has been developing boiling histotripsy Hz. During treatment planning, the entire tumor volume was covered (BH) as a non-invasive treatment for renal carcinoma (RCC) and have pre- and 2-mm spacing was used between points. Twenty-four hours after viously shown short-term changes in systemicand local cytokines and in- pFUS, tumors were harvested for histology and molecular analyses. filtration of CD8+ T-cells following BH treatment in vivo. We characterized RESULTS During pFUS, half-harmonic emissions were only detected the long-term immune response to BH RCC tumor ablation in a ratmodel. only at 8 MPa PNP in B16 tumors and at 6 and 8 MPa PNP in 4T1 tu- METHODS RCC bearing genotyped Eker rats (Tsc2 heterozygotes) were mors. Twenty-four hours after pFUS, we examined expression of randomly assigned to transcutaneous BH or FUS SHAM procedure target- intercellular adhesion molecule (ICAM), vascular cell adhesion mol- ing ~0.5 cc of RCC or non-tumor bearing normal kidney. BH was deliv- ecule (VCAM), and cyclooxygenase (COX2) in each tumortype. ICAM ered with a 1.5 MHz US-guided small animal FUS system (VIFU-2000, was elevated in 4T1 and B16 tumors between 2 and 4 MPa (P <0.05) Alpinion) operated at duty cycles of 1-2%, 10-20ms pulses, 525-600 W compared to untreated tumors (Fig. 1). We did not observe statisti- electric power. Following treatment rats were recovered, underwent serial cally significant elevations in COX2 for B16 tumors, but COX2 was el- US imaging surveillance and serial blood draws, and were survived for7, evated in 4T1 tumors treated with pFUS as 4, 6, and 8 MPa (P <0.05). 14, or 56 days. Following euthanasia, the treated and contralateral kidney, VCAM expression was not upregulated bypFUS in either tumor type tumor draining lymph nodes (TDLN), and spleen were collected. Flow-cy- at any pressure. Furthermore, H&E staining revealed wide-spread co- tometry was performed on processed tissues and blood to analyze for agulative necrosis and red blood cell extravasation in both tumor changes in circulating and local immune cell populations. typestreated at 8 MPa compared to lower pressures. RESULTS BH treatment was associated with significant alterations to the CONCLUSIONS We found that pFUS could induce molecular changes immune system within 2 weeks vs. SHAM treatment with characteristics of in B16 and 4T1 tumors that are consistent with molecular response an anti-tumor immune response (see Fig. 1). Specifically, BH was associated that can influence immune cell tropism. Interestingly, we demon- with a significant 3.4-fold (p=0.048) increase in splenic CD11c+ antigen pre- strate that there may be optimal rarefaction pressures to induce ap- senting dendritic cells. BH also produced alterations in CD4+ helper T-cell propriate molecular changes in tumors. For example, the greatest populations with a significant 1.2-fold (0.041) increase in CD4+ CD62L- increases in ICAM, that could facilitate immune cell infiltration, was CD44+ effector memory cells and near significant 1.8-fold (p=0.08) increase observed below the cavitation threshold in both tumor types. Fur- in central memory CD4+ CD62L+ CD44- cells in TDLNs. Finally, BH treat- thermore, we show that molecular responses can be fundamentally ment resulted in significant alterations in cytotoxicCD8+ T-cell populations. different across tumor types (e.g. changes in COX2 in 4T1 tumors, Specifically, BH treatment was associated with significantly increased CD8+ but not B16 tumors), that may reflect differences in the physiological CD62L- CD44+ effector memory cells in both TDLNs and spleen (3.0-fold, and functional responses. These results suggest it may be possible to p<0.01 and 2.9-fold, p=0.03, respectively) and central memory CD8+ CD62L use pFUS to modulate immune function, but that a greater under- + CD44- cells in TDLNs (6.8-fold, p <0.01) with a corresponding significant standing of pFUS molecular effectsis needed. Given the heteroge- decrease in CD8+ CD62L+ CD44+ naïve cells in TDLNs (0.4-fold, p=0.01). neous response across tumor types, that each pFUS parameters may CONCLUSIONS These data represent the first extensive analysis of the im- need to be empirically evaluated. mune response to BH tumor treatments and suggest that BH RCC ablation produces significant changes to the immune system suggestive of an anti- tumor response. Analysis of longer-term survival (56 days) is ongoing and will demonstrate if these changes are long-lived. Future studies will further evaluate the specificity of this response and whether it can improve clinic- ally relevant outcomes. Funding: The Focused Ultrasound Foundation, The Urology Care Foundation, and NIH K01EB015745 and R01CA154451. Fig. 1 (abstract O16). Flow cytometry results from tumor draining lymph nodes and spleens 2 weeks following SHAM or BH treatment Fig. 1 (abstract O15). See text for description of RCC Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 11 of 122 O17 CONCLUSIONS The results demonstrate that USMB treatments in Potentiating checkpoint inhibitor therapy with ultrasound combination with anti PD-1 therapy resulted in a significant inhib- stimulated microbubbles ition of tumor growth. The data suggest that USMB therapy may be 2 1, 2 1, 2 1, 2 Sharshi Bulner , Aaron Prodeus , Jean Gariepy , Kullervo Hynynen , potentiating anti PD-1 therapy through a T-cell dependent mechan- 1, 2 David Goertz ism. This approach is a means by which to enhance checkpoint in- 1 2 University of Toronto, Toronto, Ontario, Canada; Sunny brook Research hibitor effects without the deliterious side effects associated with Institute, Toronto, Ontario, Canada adding other immunotherapy or chemotherapy agents. Correspondence: Sharshi Bulner Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O17 O18 OBJECTIVES Immunotherapies are playing an increasingly prominent Proteomic and histological effects of pulsed focused ultrasound in role in the treatment of cancer, driven in large part by the success of the rat heart 1 1 1 1 checkpoint pathway inhibitors such as anti-PD-1 monoclonal anti- Kee W. Jang , Tsang-Wei Tu , Scott R. Burks , Bobbie K. Lewis , Joseph A. 1,2 bodies. These approaches inhibit the ability of tumors to evade the Frank immune system and boost the activity of anti-tumor Tcells. However, Radiology and Imaging Sciences, National Institutes of Health, Bethesda, potent and durable results are to date only obtained in subsets of Maryland, USA; National Institute of Biomedical Imaging and patients with certain cancers (e.g. melanoma). They are less effective Bioengineering, National Institutes of Health, Bethesda, Maryland, USA in tumors with low immunogenicity, and those with low levels of Correspondence: Kee W. Jang tumor infiltrating lymphocytes (TILs). As a result, it is increasingly rec- Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O18 ognized that combination therapies will be required. While it is known that different forms of therapeutic ultrasound can elicit a OBJECTIVES The purpose of this study was to investigate the effects spectrum of immune responses, it remains to be established if they of pulsed focused ultrasound (pFUS) exposures to the rat heart and can be harnessed to potentiate the effects of checkpoint inhibitors. evaluate the proteomic and histological changes temporally follow- In this study we examine the combination of anti-PD-1 therapy with ing sonication. ultrasound stimulated microbubbles (USMBs). Inparticular, we use a METHODS Eight to ten week old female Sprague Dawley rats were level of ultrasound exposure that is non-thermal, but elicits a signifi- imaged on a 3T MR scanner (Achieva, Philips Healthcare, USA) and cant degree of vascular injury. T2-weighted MR images were acquired with 8.9ms repetition time METHODS A murine colorectal cancer cell line (CT26.wt) was (TR), 4.5ms echo time (TE) in 1mm slice thickness through the chest employed to initate tumors subcutaneously in the hind limbs of fe- wall in sagittal plane. The MR images were used as guidance for male Balb/c mice. There were four study groups: microbubbles-only pFUS to target the left ventricular apical myocardium (center fre- (control/sham treatments), Anti PD-1 (drug-only group), USMBs (ultra- quency 1.1MHz; PNP 3MPa; 10ms bursts; 1Hz PRF; 100 sonications/ sound-only treatment group) and USMBs + Anti PD-1(combined point, 40 points to the apex, RK100, FUS Instruments, CAN). Follow- treatment group). Experiments commenced when tumors were well ing pFUS treatment, proteomic analysis was performed of sonicated established, in the range of 50-100 mm3. The immunotherapy drug myocardium lysate (n=5/timepoint) over 24 hrs along withcardiac in- used in this study is ananti-mouse PD-1 (clone: RMP1-4, Bioxcell) and jury markers from the myocardium and serum (s) using single- or was administered at a dosage of 200 μg intraperitoneally prior to multi- plexed ELISA. In order to monitor and examine the activity of ultrasound or sham treatments, and then subsequentlyadministered cardiac function, electrocardiogram (EKG) was recorded and analyzed every 3 days for a total of 5 doses. The microbubbles employed were at pre-determined time points using EKG recording module (IX- an experimental phospholipid encapsulated agent administered in ECG12, iWorx Systems, Inc., NH). pFUS treatedmyocardium (n=3) bolus form prior toinsonation. Ultrasound exposures were delivered were stained with haematoxylin and eosin (H&E) to evaluate the with a single element focused 1 MHz transducer with the entire tu- morphology. Statistical analysis was performed using ANOVA with mors within the -3dB beam width. The pulsing scheme was to send multiple comparisons to sham control with p<0.05. 50 0.1 ms long pulses (1.6 MPa peak negative pressure) spaced 1 ms RESULTS pFUS treatment to the myocardial apex induced a delayed apart, which was repeated at 10 second intervals for a duration of 3 pro-inflammatory cytokine expression with significant elevations in minutes. This low duty cycle approach avoided thermal elevations interleukin (IL) 1α,1β, 17tumor necrosis factor alpha (TNFα) inter- and has been previously established to induce vascular injury in tu- feron gamma (IFNγ) and anti-inflammatory factors including IL4, 10 mors. Antitumour effects was assessed withlongitudinal experiments along with chemo-attractant factor regulated on activation, normal T (n=5-7 per group), where tumor growth was evaluated every three cell expressed and secreted (RANTES). Significant (p<0.05) cardiac days until ethical size endpoints were reached. Two sets of acute ex- biomarkers including cardiac troponin (cTn) and N-terminal pro b- periments were conducted for all groups, where tissue (tumors, type natri-uretic peptide(NT-proBNP) were also detected in both spleens and tumor draining lymph nodes) was harvested at either serum and myocardium that would be indicative of cardiac injury 3 or 7 day time points post induction (4-6 pergroup). Tumor tis- (Fig. 1). The expression of proinflammatory cytokines including be- sue underwent flow cytomtery analysis to assess immune cell sta- came significant (p<0.05) at 12 through 24 hrs after pFUS treatment. tus. Splenic and lymphatic tissue was subject to cell count The increase in cyclo oxygenase 2 (COX2) at 24 hours would suggest analysis and ELISPOT to quantify the expression of IFN-gamma as that the molecular changes would be occurring through NFkB path- a metric of T-cell activation. way. The significantly increased levels of cardiac injury markers, NT- RESULTS It was observed that the combined treatment group had proBNP and cTnI, in both myocardium and serum were observed a significantly smaller tumor volume compared to the sham after 18 hrs following pFUS treatment. Analysis of EKG signals re- group, USMB-only group and anti PD-1group at Day 6 (p<0.001, vealed that the heart rate declined immediately after pFUS and p<0.05 and p<0.05, respectively) and at Day 9 (p<0.001, p<0.001, within 1 hour returned back to normal (Fig. 2). H&E stains did not p<0.001, respectively). Using a size dependant endpoint (>1000 show morphologic changes in the apex in pFUS treated myocardium mm3), the combined treatment group resulted in a significantly compared to sham controls (Fig. 3). longer survival time compared to MBs (p<0.01), USMB (p<0.05) CONCLUSIONS We demonstrated a reproducible and delayed mild and anti PD-1 (p<0.01). Median survival between the combined cardiac injury model using MR-guided pFUS the rat. Proteomic treatment group and individual treatments was 16 days versus 10 changes and cardiac markers wouldindicate that the evolution of in- days. Flow cytometry showed that the USMB treatments pro- flammatory response to low pressure pFUS (3MPa) starting approxi- duced elevated levels oftumor infiltrating leukocytes (TILs - CD45 mately 6 hours following pFUS would suggest that sonication is +) and that the ratio of cytotoxic T-cells (CD8+) to regulatory T causing a delayed cardiac injury. Further histological and transcrip- cells (CD45+ Foxp3+) was increased. At Day 3, the combined tomic analysis will be needed to understand the pathophysiological therapy group had significantly higher TIL and cytotoxic T-cell effects of pFUS to the heart and whether the model can be used as levels than the anti PD-1 monotherapy (p<0.05) group. a basis to evaluate myocardial contusion. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 12 of 122 OBJECTIVES Biomedical Capacitive Micro-machined Ultrasound Transducers (CMUTs) have been mainly developed for imaging pur- poses. CMUTs exhibit several advantages, such as ease of miniaturization (cell size: micron size), inherently broad bandwidth (>several MHz) and good MR compatibility. This technology also has potential for generating high intensity ultrasound, which could have an interest in developing ultrasound thermal ablation therapies. To date however, the feasibility for generating high intensity ultrasound with CMUTs has only been described in a few modeling and in-vitro studies. The use of CMUTs in continuous wave (CW) mode ofopera- tion remains challenging since it requires the development of robust cell structures and dedicated driving strategies. Here, we present an Fig. 1 (abstract O18). Proteomic analysis of pFUS treatment in a in-vivo study investigating the feasibility of generating directional rat myocardium. Pro and anti inflammatory cytokines including ultrasound-induced heating and thermal damage in brain tissue, with cardiac injury biomarkers showed delayed enhancement CMUTs designed for interstitial high intensity contact ultrasound following pFUS treatment and the increased expression lasted (HICU) applications under magnetic resonance (MR) guidance. the end of predetermined time points. Significance was METHODS Two types of CMUT prototypes were fabricated using a determined by ANOVA p <0.05 series production batch of CMUTs and a wafer bonding based process. A first intermediary CMUTprototype was made of 5 columns of 4-element 1D-linear arrays mounted on a 16 cm long, 5 mm wide rigid PCB. The elements were electrically coupled 2 by 2 within agi- ven column (element size: 2.7 mm x 0.8 mm). A final prototype was a multi-faceted CMUT-based HICU catheter incorporating ten 1D-lin- ear arrays, 32.4 mm long and 0.8 mm wide. The arrays were mounted on a cylindrical 9-French flexible catheters (20 cm long), which formed a prism-shaped 2D array for multidirectional radial ultrasound exposures. CMUT prototypes were used in a porcine model for gen- erating thermal ablations in brain tissue interstitially. Preliminary nu- merical simulations allowed identifying a range of surface ultrasound intensities (Iac) suitable for inducing thermal ablation in brain tissues Fig. 2 (abstract O18). Analysis of EKG before and after pFUS. (A) with these CMUT designs (Iac > 10 W•cm-2). CMUTs were used in CW Heart rate (HR) was immediately declined following pFUS and mode (HICU sequence: 4s ON/1s OFF, f = 7.9 MHz), and the bias and returned back to normal shortly. (B) Transient elevation of RR interval driving voltages were chosen in order to operate in the collapses confirmed the changes of HR following pFUS treatment. Significance napback regime and reach the intensity level required for thermal was determined by ANOVA ****p<0.0001 HICU lesioning. Ultrasound exposures conditions were applied through an escalation dose process (total exposure time: from 3 to 15 min, up to 10 active CMUT elements). HICU-induced thermal heat- ing generated with CMUTs was evaluated in vivo on 10 pigs and monitored under real-time multi-planar magnetic resonance therm- ometry (MRT) (Fig. 1). RESULTS Overall, directional HICU-induced temperature increases were generated in the in-vivo porcine brain with CMUTs. The heating pattern could be monitoredwith excellent time-space resolution be- Fig. 3 (abstract O18). H&E stain of pFUS treated myocardium. yond a radius of 1 mm around the CMUT device (12 MRT maps every Compared to SHAM (A), There was not difference in myocardial 1s, temperature standard deviation: ± 2.5°C). Heating patterns ex- morphology different at post 6 hrs (B), post 24 hrs (C)and post 120 tended over 1 cm from the CMUT elements within 2 min exposures. hrs (D) following pFUS. Bar = 100μm HICU exposures could be performed continuously without a water- circulating coolingsystem. After 6 min, the temperature of the brain tissue was increased locally above the 55°C threshold necessary for the creation of irreversible thermal damage (△Tmax> 35 °C). Treat- ment volumes > 1.5 cm3 could be completed within 13 min. Just after treatment, tissue changes were visible on T1- and T2-weighted anatomical images. Tissue ablation boundaries detected on T1w and O19 T2w images, respectively hypo and hyper signal boundaries, corre- CMUT transducers for interstitial MR-guided high intensity lated well with the 55°C isotherm boundaries (MRTmaps). Several ultrasound heating: in-vivo feasibility in a pig brain model 1,2 1,2 1,2 contrasts were observable on MR images within the lesion area, William Apoutou N'Djin , Jeremy Vion Loïc DAUNIZEAU , 2 1, 3 4 which were consistent with previous studies reporting the presence Christopher Bawiec , Guillaume BOUCHOUX , Nicolas Sénégond , 1,2 3,5 of coagulation andliquefaction necrosis in brain after high intensity Jean-Yves Chapelon , Alexandre CARPENTIER 1 2 ultrasound exposures. Gross sample and histological analyses con- LabTAU, INSERM, U1032, Lyon, Rhone-Alpes, France; Univ Lyon, firmed the presence of brain tissue coagulations. Université Lyon 1, Lyon, F-69003, France; CarThera Research Team, Brain and Spine Institute, Paris, France; Vermon SA, Tours, France; CONCLUSIONS The feasibility of HICU therapy with CMUTs has been Department of Neurosurgery, Assistance Publique, Hopitaux de Paris, established in vivo. Further investigations are ongoing to improve the Pitie Salpetriere, Paris, France robustness of the CMUTdevices and increase the treatment volumes. Correspondence: William Apoutou N'Djin This project was supported by CarThera, the French National Research Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O19 Agency (ANR, 2010) and Single InterministerialFund (FUI, 2013). Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 13 of 122 reconstruction algorithm implemented in a ultrasound beamformers with digital parallel processing units; 4) generating the HIFU beam image. The experimental set-up included a mono-element HIFU spherical transducer (emitter, focal distance: 90 mm, working fre- quency: 2.4 MHz) and a commercial ultrasound imaging probe (re- ceiver, model: L7-4, ATL-Philips) for collecting the emitted HIFU signals. In this experiment, the emitter was aligned with the receiver face to face each other. For the wave field reconstruction, an ultrafast ultrasound scanner (Vantage 256, Verasonics) was used for the signal recording and processing via plane-wave beamforming. Evaluations: The feasibility of the method was evaluated using either the time-re- versal reconstruction or the beamforming reconstruction, with and without HIFU beam aberrations. First, in the experiment in water without HIFU beam aberrations, performance of the method was demonstrated in comparison with the reference field obtained by nu- merical calculation of the forward propagation using the Rayleigh in- Fig. 1 (abstract O19). In-vivo interstitial HICU heating induced in a tegral and hydrophone scanning. Then, in the experiment with HIFU porcine brain with CMUTs under MR temperature monitoring beam aberrations induced by heterogeneous hydrogel, HIFU beam visibility was evaluated with referred to the HIFU pressure field mea- sured by hydrophone scanning. RESULTS In a lossless medium (water), qualitative agreements were found between the amplitude of the reference HIFU field from numerical calculation (Fig. 1a) andthe HIFU fields obtained O20 from real experimental acquisitions after time-reversal and Real-time HIFU beam imaging using beamforming in an beamforming reconstructions (Fig. 1b and c). In the case of a ultrasound scanner heterogeneous medium including multiple bubbles and cracks, 1 2, 4 3 Kazuhiro Matsui , Françoise CHAVRIER , Takashi Azuma , Ichiro the HIFU beam was aberrated and split by the ultrasound 1, 5 2, 4 2,4 Sakuma , William Apoutou N'Djin , Rémi Souchon propagation through the aberrators (Fig. 2). The aberrated HIFU 1 2 Engineering, The University of Tokyo, Tokyo, Japan; LabTAU, INSERM fields obtained by time-reversal reconstruction (Fig. 2a and d) unité 1032, Lyon, France; Medicine, The University of Tokyo, Tokyo, and using the beamforming reconstruction (Fig. 2b and e), 4 5 Japan; Univ Lyon, Université Lyon 1, Lyon, France; Medical Device closely matches the reference field obtainedby hydrophone Development and Regulation Research Center, Tokyo, Japan measurement (Fig. 2c). Correspondence: Kazuhiro Matsui CONCLUSIONS The results indicated the feasibility of the method, al- Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O20 though reconstruction errors were observed, arising from the finite aperture size of the probe or from large path lengths from the OBJECTIVES High Intensity Focused Ultrasound (HIFU) therapies can source. Despite this limitation, the proposed method may be accept- currently be assisted by medical imaging for guidance or monitoring able as an easy, simple, and real-time HIFU beam imaging techni- of some tissue characteristics (e.g. anatomical structures, physio- queused for therapy guidance if the HIFU emitter and its receiver logical parameters), but no visualization of the HIFU beam itself is can be located in a proper face-to-face orientation. currently provided for assisting the treatment targeting. In biological tissues, however, the HIFU beam may be strongly attenuated, or may even focus outside the desired target region, for example due to un- controllable experimental parameters (acoustic coupling at the inter- face of the ultrasound source, anatomical barriers such as bones, gas), or due to multiple tissue layers having different ultrasoundpro- pagation properties (e.g. muscle, skin or fat). These experimental pa- rameters are difficult to quantify precisely a priori, and cannot be completely accounted for during calculation of the focusing parame- ters.There is therefore a risk of applying the HIFU energy and thus creating a therapeutic effect (e.g. thermal lesion) at a wrong position inside the medium, which potential deleterious consequences on the efficacy, accuracy, and safety of the treatment. The object of this study is to provide a method for imaging, in real time, the HIFU beam inside an acoustically propagative medium using beamforming in an ultrasound scanner.Novelty and advanta- gesThe novelty of the method can be found in its implementation of beamforming in an ultrasound scanner, which is analogous to the backward reconstruction using time reversal. The advantage of this method is its real-time imaging capability owing to digital parallel processing of the scanner. Further advantage is simplicity of the im- aging system without requiring additional equipment aside from an Fig. 1 (abstract O20). HIFU amplitude field in a lossless medium ultrasound scanner and an ultrasound array serving as a time-reversal (water). The horizontal and vertical axes are respectively x and zaxes. mirror. Each image is normalized by maximum amplitude. (a)Numerical METHODS Imaging technique:The method for imaging in real time calculation via forward propagation using Rayleigh integral. (b) the HIFU beam comprising steps of: 1) emitting a HIFU beam in the Experimental acquisition using backward reconstruction using tissues; 2) measuring voltage signal waveform values at a reception Rayleigh integral. (c)Experimental acquisition using backward region using an ultrasound array probe serving as a time-reversal reconstruction using beamforming mirror; 3) calculating acoustic field values in the tissues using the Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 14 of 122 RESULTS MEUS treatment for 5 min successfully shut down or re- duced the blood perfusion of rabbit livers. The average peak inten- sity (PI) of CEUS dropped significantly from 84.0±4.3% to 49.3±5.1%. The average volume of ablation zone (4.55±0.83 cm3) treated by MEUS plus RFA combination was significantly bigger than that of RFA treatment alone (1.63±0.29 cm3) (Fig. 1). The levels of serum ALT and AST did rise after both of two treatments but recovered to normal eight days later (Figs. 2, 3). CONCLUSIONS Liver RFA can be greatly enhanced by combining MEUS pretreatment. Fig. 2 (abstract O20). HIFU amplitude field with aberrations obtained experimentally. Each image is normalized by maximum amplitude. (a) Wideview image obtained by backward reconstruction using Rayleigh integral. (b) Wideview image obtained by backward reconstruction using beamforming. (c) Reference image obtained by hydrophone measurements in the ROI of 8 × 20 mm2. (d) Reconstructed image via Fig. 1 (abstract O22). See text for description Rayleigh integral in the ROI. (e) Reconstructed image via beamforming in the ROI O21 Low-intensity ultrasound extends lifetimes of transplanted mesenchymal stromal cells in murine muscle Scott R. Burks, Matthew Nagle, Saejeong Kim, Joseph A. Frank Radiology and Imaging Sciences, NIH Clinical Center, Bethesda, Maryland, USA Fig. 2 (abstract O22). See text for description Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O21 Correspondence: Scott R. Burks Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O21 This abstract is not included as it has already been published: Burks S, Nagle M, Kim S, Milo B, Frank J. A90 Low-intensity ultrasound prolongs lifetimes of transplanted mesenchymal stem cells. J Ther Ultrasound. 2016; 4(Suppl 1):31. Available from: https://jtultrasound.- biomedcentral.com/articles/10.1186/s40349-016-0076-5 O22 Fig. 3 (abstract O22). See text for description Impact of microbubble-enhanced radiofrequency ablation of rabbit liver Zhong Chen, Xueyan Qiao Department of Ultrasound, Xinqiao Hospital, The Third Military Medical University, Chongqing, China Correspondence: Zhong Chen O23 Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O22 In vitro and in vivo investigation of high intensity focused ultrasound (HIFU) hat-type ablation mode OBJECTIVES To investigate the synergistic effect of combining micro- Hongya Dai bubble-enhanced ultrasound (MEUS) and radiofrequency ablation Biomedical Engineering, Chongqing Medical University, ChongQing, (RFA) on normal rabbit liver. ChongQing, China METHODS Eighteen surgically exposed rabbit livers were treated Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O23 with MEUS immediately followed by RFA. The other 18 livers were treated with RFA alone. The therapeutic ultrasound of MEUS was op- This abstract is not included as it has already been published: erated at a pressure amplitude of 4.3 MPa and a duty cycle of 0.22%. Dai H, Chen F, Yan S, Ding X, Ma D, Wen J, Xu D, Zou X. In Vitro and Contrast-enhanced ultrasound (CEUS) was used to evaluatethe liver In Vivo Investigation of High-Intensity Focused Ultrasound (HIFU) circulation. Twenty livers were removed for volume measurement of Hat-Type Ablation Mode. Med Sci Monit. 2017; 23:3373-3382. Avail- ablation and the rest livers were also harvested for histological exam- able from: https://www.medscimonit.com/abstract/index/idArt/ ination 24h after treatment. Serum ALT and AST levels of 10 rabbits 902528 DOI: 10.12659/MSM.902528 were examined to monitor any changes of liver function. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 15 of 122 O24 The skin interface was identified in imaging to be approximately Image-based predicition of focusing gain in situ using dual-mode 34mm. Based on the acoustic power delivered and using re- ultrasound arrays ported numbers of acoustic properties of muscle the calculated Brogan T. McWilliams, Dalong Liu, Emad S. Ebbini heating rate was 9629°C/s. This is consistent with imaging data Electrical Engineering, University of Minnesota, Minneapolis, Minnesota, of the delivered therapy exhibiting a bubbles from boiling condi- USA tions within <100 ms. Correspondence: Brogan T. McWilliams CONCLUSIONS An image-based method for real-time estimation of Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O24 the focusing gain using DMUA imaging data was validated in vitro at multiple frequencies. The method has been applied for the estima- OBJECTIVES We have previously described a dual-mode ultrasound array tion of the focusing gain in situ from real-time DMUA imaging data (DMUA) system for therapeutic image-guided focused ultrasound (IgFUS) in vivo. As described, STF imaging of the phantom slab allowed for applications. This system has been shown to provide closed-loop spatio- measuring the beam dimensions and the FUS interaction with the temporal control of FUS beams with 2 millisecond and submillimeter time tissue and could be used in future studies to extract details of an in- and spatial resolutions, respectively. With appropriate feedback (e.g. homogeneous medium to provide accurate estimates of the focusing temperature) this system offers the promise of delivering precision HIFU gain (or heating rate). Further validation of the calculated heating therapy, which would improve the safety and efficacy of IgFUS proce- rate will be performed in vitro and in vivo by measuringtemperature dures. A key performance parameter for HIFU therapy is the focusing gain, with thermocouples in the vicinity of the focus. which is a characteristic of the transducer (array). The imaging capability of the DMUA allow for the estimation and characterization of the precise imagingpathof the HIFU beam throughthe bolusand interveningtissue to HIFU target. The objective of this paper is to validate an image-based approach for the estimation of the heating rate in situ utilizing simplified propagation models and calibration power measurements. METHODS In this study, we used a 3.5-MHz, 64-element DMUA proto- type with concave aperture (40-mm radius of curvature with f# = 1). This prototype was used for generating the HIFU beam as well as col- lecting the corresponding imaging data. Figure 1 shows the in vitro setup used to provide a validation for the image-based approach. In particular, an acoustic power meter (Omega, Ohmic instruments Easton, MD) was modified to allow the measurement of the insertion loss due to a slab of tissue-mimicking phantom between the DMUA and the tip of the cone (treated as the target). The TM phantom was fabricated from animal skin bovine gelatine, graphite,1-propanol, glutaraldehyde, and deionized water. Absorption of the phantom is predominately due Fig. 1 (abstract O24). a) shows a digital photo of the power meter to the presence of graphite and was determined to be 0.6 and 1.0dB/ with the 3D printed support bracket and slot placed into the water cm/MHz for two 4-mm disk-shaped slaps. Two modes of the imaging tank. b) shows a cross section along the center toshow orientation were used to characterize the FUS beam propagation through the tis- of the DMUA and phantom with respect to the cone. the length d is sue: 1) Synthetic-aperture(SA) imaging, which provides larger field of adjustable for disk1 and 2 this length was set to 4 mm view (FoV) to characterize the propagation medium, and 2) Single- transmit focus (STF), which provides specific feedback about the inter- actions between the FUS beam and the tissue in its path to the target. The STF imaging is performed using the same beamforming parame- ters of the intended therapeutic HIFU beam, but at diagnostic levels and with sub-microsecond pulse duration. HIFU was applied at 4 differ- ent frequencies (2.4 to 4.2 MHz in stepsof 0.6 MHz). HIFU shots of 1-sec durations were used and repeated 4 times. SA and STF images were collected before, during and after the application of therapeutic HIFU. The STF frame rate was 400 fps, which was helpful to fully characterize the incidence of cavitation and/or instability in the power measurement. RESULTS Figure 2 shows SA and STF images with and without the phantom slab in place. From Fig. 2.b an approximated dimesion of the beam at the geometric focus was determined by the -6 dB con- tour to be approximately 0.3 x 0.5 mm. Furthermore, using a phan- tom that results in speckle reflections shown in Fig. 2.d) the beam dimensions can be observed as it traverses through the phantom. Figure 3.c) shows the thickness of the phantom to be about 4mm and ending at approximately 39mm. Figure 3 shows the average (n=4) precentage of power absorbed by the respective phantom disks and validates the approach described in methods at variousfre- Fig. 2 (abstract O24). (a) SA image of the power meter without the quencies. The calculated and measured power absorption show simi- phantom slab. (b) STF image of the power meter without the lar trends with respect to frequency. Figure 4 shows an SA image of phantom slab. (c) SA image of the power meter with the phantom a rat's left carotid artery in vivofrom an image-guided targeting ex- slab. (d) STF image of the power meter with the phantom slab periment performed using the DMUA prototype described above. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 16 of 122 the gaps between adjacent lesions may have a seeding effect, leading to further growth of malignancies due to the untreated sections. To avoid lesion interactions, several authors suggested a pre-defined time delay betweenlesions or change in exposure parameters. The former results in unnecessary treatment delays while the latter involves cap- acious real-time computations for optimizing the treatment (dynamic- ally computing thermal dose) as well as remotely and frequently switching on/off high power equipment. In order to overcome these deficiencies, we propose a method for determining the optimal lesion arrangement within any arbitrary tumor size and shape, based on an extension of the bubble packing algorithm first described by Shimada in 1995 [1]. The original algorithm was extended to allow lesions to take any arbitrary position andorientation within the specified tumor volume, and evaluated on spherical and ellipsoidal tumor models. METHODS The bubble packing algorithm first described by Shimada was extended to the case of arbitrary tumor position and orientation. The existing approach utilizes a model in which overlapping lesions apply forces to adjacent lesions until an equilibrium is reached. Additional forces are applied at the boundary of the tumor to ensure lesions remain within the tumor volume. In this work, the model has been extended such that Fig. 3 (abstract O24). Percentage of power in therapy shot absorbed in addition to the force, a torque is applied resulting in changes in lesion by phantom slab. Disk 1 and 2 corresponds to 0.6 and 1.0 dB/cmMHz orientation and allowing the optimization of lesion configuration in all six respectively and are both a thickness of 4mm dimensions. Initial evaluation of the technique was performed based on two simplified tumor models. While calculating and fitting the lesions, we assumed our HIFU probe focus as a3×7×3mm ellipsoidal volume and a 20mm diameter spherical target tumor volume, and a 10×5×5 mm ellips- oidal volume. As lesion placement may be constrained to a single acous- tic window located some distance from the organ, the case of fixed lesion orientation was also considered as well as the free orientation and trad- itional raster approach. During raster lesion placement, lesions were gen- erated at evenly spaced intervals throughout the tumor and surrounding volume; those lesions of which the centroid fell outside the tumor volume were then excluded. To allow direct comparison of the approaches, the number of lesions in the optimized cases was set to the same as that uti- lized for the raster case. In each case Monte-Carlo were utilized to esti- mate the total portion of tumor tissue destroyed, as well as the total volume overlap and volume of healthy tissue (i.e. that outside of the de- fined tumor region) ablated. Points were randomly generated within a do- main covering twice the tumor volume, with those falling within the targeted region but none of the lesion sites considered as untreated, those within both the tumor volume and one (or more) lesion volume considered as treated, while those outside of the tumor volume but within one or more of the lesions considered as ablated healthy tissue. The optimization process was completed 10 times for each configuration. RESULTS The modified bubble packing approach was successfully imple- mented in C++ and all described evaluation cases successfully performed. The achieved results demonstrate that the introduction of the bubble Fig. 4 (abstract O24). SA imaging of therapy plane with carotid packing approach leads to an improvement of approximately 10% in total body in vivo treated volume as compared to the raster approach, as well as a decrease in ablated healthy tissue of up to 20% (Fig. 1, Table 1). Slight differences in the free and fixed orientation cases, specifically increases in overlap and decreases in ablation of healthy tissue were observed with the intro- O25 duction of orientation into the optimization framework. An automatic approach to lesion planning for robotic HIFU CONCLUSIONS This work described the development and evaluation of a Tom Williamson, Scott Everitt, Ranjaka De Mel, Sunita Chauhan general automated method for the treatment planning of lesion positions dur- Mechanical and Aerospace Engineering, Monash University, Melbourne, ing HIFU. Theapproach allows the determination of optimal lesion locations Victoria, Australia and orientations, and can be applied to arbitrary tumor shapes and sizes as Correspondence: Tom Williamson well as arbitrary lesion configurations. Evaluation of the approach on spherical Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O25 and ellipsoidal tumor models demonstrated the feasibility of the approach. OBJECTIVES High intensity focused ultrasound (HIFU) based ablation has been found to be predictable and successful for treatment of ‘car- cinoma in situ’ in variousorgans. At this stage, the neoplasm is generally spherical in shape while, conventionally, focused ultrasound (FUS) treat- ment involves ‘raster’ scanning the formation ofthe HIFU lesions in the ROI, an approach that will generally not conform to the spherical tumor geometry. This may lead to two major undesirable effects: large gaps or overlaps at the tumor margins (physical spacing isn’t optimized) and between individual lesions, or significant ‘lesion-to-lesion’ interaction creating uncertainty in the shape, size and extent of the subsequent le- Fig. 1 (abstract O25). See text for description sions due to the remnant effect of previously laid lesions. Furthermore, Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 17 of 122 Table 1 (abstract O25). Summarized results of algorithm evaluation of 3.8 ± 1.5 mm. Ultrasound images revealed a hyperechoic region that was well correlated with the macroscopic analyses of the HIFU le- sions. Necrosis of placental tissues exposed to HIFU was confirmed with macroscopic and microscopic analyses. No significant variation in ma- ternal and fetalparameters was observed during HIFU exposure. Histo- logical examination demonstrated coagulative necrosis within the core of the lesion. Terminal villi were split with trophoblast desquamation, mesenchymal involution and fibrinoid deposits containing fragmented red blood cells. Congestion and hemorrhagic villitis were seen at the- border of the lesion. No damage was seen in the chorionic vessels wall or endothelial cells. There was no inflammation. CONCLUSIONS This study demonstrates in a monkey model of preg- nancy the feasibility of HIFU treatment applied to the placenta for potential application to treattwin-to-twin transfusion syndrome. We report here the first use of a completely non-invasive treatment of the placenta in pregnant animals. The primary aims of this study O26 were to assess the efficacy and safety of this technique for the preg- Extracorporeal high-intensity focused ultrasound treatment of the nant monkey and fetus. The presence of well-defined and controlled placental unit: in vivo study using a monkey model of pregnancy lesions in the placenta without complications was encouraging. Five 1 1 1 2 David Melodelima , Jonathan Caloone , Anthony Kocot , Cyril Huissoud HIFU lesions were created in the anterior placenta, and one was cre- 1 2 LabTAU - U1032, INSERM, Lyon, Rhône Alpes, France; CHU Croix ated in the posterior placenta. The insonification of the posterior pla- Rousse, LYON, France centa was possible without any complications, but the risk to the Correspondence: David Melodelima fetus is much more important. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O26 OBJECTIVES Fetal surgery represents an interesting approach for sev- eral fetal diseases, particularly for the treatment of twin-to-twin trans- fusion syndrome (TTTS). Although fetoscopy increases survival rate, it O27 is also considered invasive and responsible for fetal and maternal Reasons of different therapeutic efficacy of HIFU ablation for complications that can affect neonatal outcome. These complications uterine fibroids with different MRI-T2W signal: thermal, acoustic are partially associated with the opening of the amniotic cavity. A and histopathological properties study completely non-invasive treatment that could occlude deep anasto- Faqi Li, Haoran Huang, Jianbo Ran, Zonggui Chen, Man Luo, Fei Li, moses would prevent the risks of invasive fetoscopy while offering Qi Wang, Jie Xu, Huan Liu, Zhibiao Wang the potential for more effective therapy. The efficacy of high-intensity College of Biomedical Engineering, State Key Laboratory of Ultrasound focused ultrasound (HIFU) is clinically proven for non-invasive tissue Engineering in Medicine Co-founded by Chongqing and the Ministry of ablation. We previously developed a toroidal HIFU transducer that Science and Technology, Chongqing Key Laboratory of Ultrasound in enables the destruction of large tissue volumes. In a recent study, we Medicine and Engineering, Chongqing Medical University, Chongqing, demonstrated the ability to induce lesions in human placenta using China this toroidal-shaped transducer without damaging intervening tissues Correspondence: Faqi Li within an ex vivo model. The effectiveness of this HIFU device ap- Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O27 plied to the perfused placental unit must be studied in a preclinical animal study under conditions similar to those in humans before OBJECTIVES To explore the causes of the differences in the thera- starting a clinical trial. Here, we report a feasibility study using a peutic efficacy of HIFU ablation for uterine fibroids with different T2- monkey model of pregnancy. The 3 objectives of this work were (i) weighted MRI signals from the perspective of thermal, acoustic and to evaluate the feasibility and reproducibility of HIFU lesions in the histopathologic characteristics. placenta of pregnant monkeys in vivo; (ii) to study the lesions using METHODS Totally 47 uterine fibroid specimens were collected after sur- ultrasound images, gross pathology, and microscopy; and (iii) to gery. According to the preoperative T2-weighted MRI signal, the uterine evaluate local andgeneral tolerance to the treatment. fibroids were classified into four categories, which were hypo-intense, iso- METHODS A toroidal HIFU transducer working at 2.5 MHz and com- intense, heterogeneous hyper-intense and homogeneous hyper-intense. posed of 32 ring-shaped emitters was used. An ultrasound probe work- The T2-weighted MRI signal of preoperative uterine fibroid was analyzed ing at 7.5 MHz was placed in the center of the HIFU transducer. The quantitatively to the signal intensity. Part of the uterine fibroids speci- imaging plane was aligned with the HIFU acoustic axis. The acoustic pa- mens were undergone the sound velocity, sound attenuation, specific rameters used during HIFU exposures were selected according to pre- heat capacity and thermal conductivity measurement, and others were liminary simulations taking into account the attenuation coefficient of used to analyze the content of tissue smooth muscle cell (SMC) and col- placentas, measured previously. In vivo experiments were then per- lagenous fiber (CF) by histopathological observation. HIFU irradiation was formed. Eight pregnant monkeys were exposed to HIFU treatment after made in the uterine fibroids specimens. Besides, with the clinical applica- anesthesia. Lesions on placental tissues were performed non-invasively tion and follow-up survey, the therapeutic effect of HIFU treatment for by placing the HIFU probe on the skin. Fetal and maternal parameters, the uterine fibroids was analyzed retrospectively. such as maternal heart rate, fetal heart rate, and subcutaneous and RESULTS There were the differences among the sound velocity, intra-amniotic fluid temperature, were recorded during HIFU exposure. sound attenuation, specific heat capacity and thermal conductivity of A C-section was performed immediately after insonification to extract the four groups of uterine fibroids, as well as the content of tissue the placenta, inspect the fetus and inspect the maternal abdominal smooth muscle cell and collagen fiber. The therapeutic efficacy has cavity. Placental HIFU lesions were studied using ultrasound images, significant differences in four groups of uterine fibroids, the groups gross pathology and histology. from the hypo-intense to homogeneous hyper-intense, the difficulty RESULTS The average gestational age of the animals was 72±4 days. of HIFU ablation have been gradually increased. There were correla- Thirteen HIFU exposures were performed on placentas. The parame- tions between the signal intensity, acoustic, thermal properties, histo- ters used in this study wereacoustic powers of 65, 80, 110 and 120 W pathological features and therapeutic efficacy of HIFU ablation for applied for 30, 15, 20 and 20 seconds, respectively. This gradual in- the uterine fibroids of different T2-weighted MRI signal. crease in the total energy delivered was used to determine aset of CONCLUSIONS There are some differences in the sound velocity, parameters to create reproducible lesions in the placentas without sound attenuation, number of SMC and CF content among the uter- any complications. Six placental lesions were observed with average ine fibroids of different MRIT2WI signal, which may be one of the im- diameters of 6.4 ± 0.5 mmand 7.8 ± 0.7 mm and an average depth portant factors to different therapeutic efficacy of HIFU ablation. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 18 of 122 (This study was supported by the National Natural ScienceFounda- avoid overexposure due to the heterogeneity of the tissues in the tar- tion of China (Grant Nos. 81127901, 11574039, 11274404)) get volume with significant variation in local absorption. open-loop control results, not shown, often resulted in either overexposure or underexposure depending on the value of the initial HIFU intensity. O28 The results also demonstrated the safety of the procedure as all the ani- Image-guided ablation of the carotid body in vivo: a potential mals tolerated this procedure well (over 32 animals at thetime of writ- noninvasive treatment for hypertension ing). Overall, the results show the feasibility of delivering prescribed 2,1 1 1 Dalong Liu , Raj Aravalli , Emad S. Ebbini levels of HIFU exposure to extremely small neurovascular structures Electrical and Computer Engineering, University of Minnesota, with potential application in the treatment of hypertension, targeted Minneapolis, Minnesota, USA; Siemens, Seattle, Washington, USA neuromodulation and other image-guided interventions. Based on the Correspondence: Dalong Liu results reported herein, a proof-of-concept study in 56 normotensive Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O28 and spontaneously hypertensive rats is currently under way. OBJECTIVES The carotid body (CB) is a neurovascular structure lo- cated near the carotid bifurcation. It is an arterial chemoreceptor, which produces neural output reflecting the partial pressure of O2 and CO2 as well as pH and temperature. The CB chemoreflex-evoked sympatho-excitatory responses are known to be enhanced in both human patients and animal model of hypertension. The main aim of this study is to establish the feasibility and safety of localized abla- tion of the carotid body usingimage-guided focused ultrasound (IgFUS) in normotensive and hypertensive rat models. METHODS Normotensive and hypertensive rats (275 - 330 gm) were treated using our dual-mode ultrasound array (DMUA) system for IgFUS using IACUC-approved protocol. In each experiment, the rat was placed in supine on stereotaxic stage with a heating pad placed in between to regu- late body temperature. The hair was shaved and removed using depilatory cream. The DMUA was positioned using a 3 stage motor under real-time imaging guidance. Synthetic-aperture (SA) Imaging was performed to lo- cate the carotid artery bifurcation based on a 3D scan of the treatment volume containing the target CB. The vessel pulsation was used to track the common (CCA), external (ECA) and internal (ICA) arteries in each plane. Once the carotid bifurcation plane was determined, four (4) treatment planes were definedoffset by 0.5, 1, 1.5 and 2 mm from the bifurcation in the caudal-to-rostral direction. In each plane, IgFUS was delivered at 2 - 6 Fig. 1 (abstract O28). See text for description locations around the ECA using a closed loop dose control (CLC) algorithm based on real-time DMUA imaging feedback (described previously at ISTU2015). Briefly, the echogenicity changes indicative of cavitation and/or tissue boiling are detected using DMUA imaging at up to 500 frames per second. These changes are used to adjust the HIFU exposure from frame to frame to avoid over-exposure to the target. The locations of the HIFU shots were chosen to assess the nature and extent of damage due to a "prescribed HIFU dose" under CLC in and around the target volume, in- cluding muscle and connective tissues. For each shot, the initial in situ HIFU intensity was between 5 - 10 kW/cm2, but this was quickly adjusted down by the CLC upon detection of the echogenicity change in beam- formed DMUA imaging data. The exposure time was 0.5 sec for each shot with minimum 40 ms at the initial intensity to ensure the echogenicity change is due to HIFU-induced change. Animals were survived for 48 - 96 hours after the treatment to allow for histological evaluation of HIFU-in- duced lesions (H&E) and the affected neural structures (Chromagranin A). Histology sections corresponding to the DMUA imaging slices during treat- ment were produced (50 μm per section covering the treatment volume.) RESULTS DMUA imaging data was collected before, during and after each HIFU shot at frame rates between 200 - 500 fps to document the lesion detection capabilitiesof the real-time guidance system. The system also logged the instantaneous intensity values used by the Fig. 2 (abstract O28). See text for description CLC algorithm. Figure 1 shows an example guidance and monitoring display from our DMUA system. The false color overlay represents the echogenicity change due to a HIFU shot placed just to the left of the right ECA nearthe bifurcation. The spatiotemporal maps (tem- poral-axial and temporal-lateral) demonstrate the high degree of O29 localization of the echogenicity change in DMUA imaging. The line Computer-aided transcranial ultrasound for time-efficient blood- profiles show the echogenicity change (in dB) at three points corre- brain barrier opening in primates 1 1 1 1 , Christian Aurup ,Carlos J. SierraSánchez , Julien Grondin , sponding to the target (red), vessel wall (blue) and skin (orange). Fig- Shih-Ying Wu 1 2 1,3 Wenlan Zheng , Vincent Ferrera , Elisa E. Konofagou ure 2 shows a typical H&E histology slide showing several discrete HIFU-induced shots (arrow heads) around the ECA. Biomedical Engineering, Columbia University, New York, New York, USA; Neuroscience, Columbia University, New York, New York, USA, CONCLUSIONS The results confirmed our ability to precisely control Radiology, Columbia University, New York, New York, USA the characteristics of HIFU-induced lesions very close to the vessel walls Correspondence: Shih-Ying Wu without damaging the endothelium, i.e. no danger of vessel perfor- ation. Furthermore, the results confirmed that CLC was essential to Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O29 Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 19 of 122 OBJECTIVES Focused ultrasound (FUS) with microbubbles holds great OBJECTIVES One third of the worldwide disease burden is caused by promise to noninvasively treat central-nervous-system diseases such as brain diseases, such as dementia, Parkinson’s and brain cancer. Des- Parkinson’sand Alzheimer’s disease through blood-brain barrier (BBB) pite efforts to develop new treatments, there remains no cure for opening and enhanced delivery of pharmaceuticals. While research has these diseases. A major reason for the lack of therapies is that most shown the need of repetitive application for treatment efficacy, currently drugs cannot enter the brain, because they are blocked by the there exists no clinical system designed to satisfy the requirement. In fact, blood-brain barrier (BBB). A promising solution uses focused ultra- the commonly-used magnetic resonance-guided FUS (MRgFUS) system sound and microbubbles to noninvasively and locally increase the hinders the re-application owing to its low speed, flexibility, and high cost BBB permeability so that drugs can enter the brain. This technology due to the dependence upon the MRI scanner. Therefore, in this study, a has successfully delivered a wide range of drugs, such as antibodies time-efficient transcranial FUS and monitoring system has been devel- and nanoparticles, and is currently being tested in human patients oped using computer-aided neuronavigation without requiring an online for the treatment of brain cancers. Despite encouraging results, the MRI, with the protocol from in silico planning, real-time targeting and current technology generates a poor drug distribution and too much monitoring, to post-treatment assessment evaluated in non-human pri- toxicity and damage for use in other brain diseases, such as demen- mates in vivo in preparation for clinical trials. tia, which affects 46 million people worldwide. Our group has re- METHODS An arm-free, real-time neuronavigation system designed for cently shown in vitro that these limitations may be due to the primates (both monkeys and humans) was customized to be used for conventional ultrasound sequences used to disrupt the BBB. These FUS application and monitoring purposes. The ultrasound system con- sequences consist of long-pulses emitted at a slow rate and generate sisted of a FUS treatment unit controlled by a customized program in a mixture of both desired and undesired cavitation activities. We Matlab with a single-element, 0.5-MHz FUS transducer triggered by a have recently developed and tested a new low pressure rapid short- function generator after 50-dB amplification, and an acoustic monitor- pulse (RaSP) sequence in vitro, designed to promote the desired cavi- ing unit with a programmable acoustic signal acquisition system and tation activity in the correct locations (e.g., capillaries). This new se- an array of acoustic detectors synchronized with the FUS system for quence is evaluated here for its ability to improve the in vivo real-time passive acoustic mapping and storage of the entire acoustic efficiency and safety of ultrasound-mediated drug delivery to the signals. Both the FUS and acoustic mapping were guided with the neu- brain. ronavigation system during the FUS procedure. The system was tested METHODS Rapid short-pulse (RaSP) sequences consist of short in 4 rhesus macaques with BBB opening in both sedate (animal under pulses separated by off-time intervals in the range of μs. In vitro anesthesia lying prone on an operating table) and awake setups (ani- studies have shown that these sequences prolong the lifetime of mal trained to sit in a customized chair), and the demonstrated proto- microbubbles and increase their mobility during the off-time in- col covered from in silico preplanning and simulation, real-time tervals, allowing lightly stimulated microbubbles to freely distrib- targeting and acoustic mapping guided by neuronavigation, to post- ute throughout the capillaries, which are the target treatment assessment of BBB opening effectivenss and safety in the microstructures of drug transfer. The better spatial and temporal MRI including contrast-enhanced T1-weighted imaging, T2-weighted distribution of microbubble activity allows a more uniform en- imaging, and susceptibility-weighted imaging (SWI). hancement of the BBB permeability and therefore a more uni- RESULTS Simulation revealed inter-animal variation (21%) due to varying form distribution of the delivered drug. We tested here whether skull density and thickness and can be compensated before sonication to RaSP sequences used in vivo in C57BL/6 mice would improve the ensure treatment reproducibility. In the in vivo experiments, for the first efficiency and safety of brain drug delivery. We compared our time the noninvasive FUS treatment was achieved within 30 min outside RaSP sequence (peak-negative pressure: 400 kPa; pulse length the MRI system for primates, and targeting flexibility allowed BBB open- (PL): 5 cycles; pulse repetition frequency (PRF): 1.25 kHz; burst ing in both the peripheral cerebral cortex and deeply-seated subcortical length: 10 ms) to the current gold standard, conventional se- structures with the use of a free-guide arm and the inflatable bladder sys- quenceat thesameacousticpressure(PL: 10,000cycles; PRF: 0.5 tem of the FUS transducer to couple with the scalp. Moreover, the Hz; burst length: 10 ms). Fluorescently-tagged (Texas Red) 3 kDa achieved targeting accuracy were proximal to the predicted 2-mm limit dextran and microbubbles were intravenously injected in mice in simulation. The accuracy in the awake animal setting (3.0 mm) was while sonicating the left hippocampus with a 1 MHz focused found to be comparable to the sedate animal setting (3.2 mm), which ultrasound transducer. A 7.5 MHz passive cavitation detector cap- was higher compared to the frame-based stereotaxis due to the improve- tured the microbubble acoustic emissions. The relative dose and ment of lateral positioning of the animal, and the focal shift in the acous- distribution of the drug were quantified by calculating the nor- tic beam path was due to the skull distortion. On the other hand, real- malised optical density (NOD, average increase in fluorescence in time cavitation mapping was performed with neuronavigation guidance the targeted area normalised by the control) and the coefficient during the entire sonication, with the frequency spectra data showed a of variation (COV, standard deviation over the average fluores- dramatic increase of the cavitation signal (harmonics and ultraharmonics) cence intensity in the targeted region). Safety was assessed by after injecting microbubbles. The acoustic mapping provided real-time haematoxylin and eosin (H&E) histological staining. spatial monitoring during the sonication and successfully confirmed the RESULTS Despite emitting 150 times less acoustic energy, RaSP location of BBB opening. Finally, in all the experiments performed, no sequences delivered a dextran dose of the same order of mag- acute damage such as hemorrhage (SWI) or edema (T2-weighted im- nitude as the long-pulse sequences (Fig. 2). Moreover, the drug aging) was detected upon radiologic examination 2 h after sonication. distribution was significantly more uniform using RaSP se- CONCLUSIONS This computer-aided system developed enabled rapid quences (Fig. 1), as indicated by the coefficient of variation (Fig. and flexible transcranial FUS applications for primates outside of the 2). Compared to conventional sequences, RaSP sequences pro- MRI scanner without the use of a stereotactic frame. It will greatly fa- duced less arterial effects, such as dextran accumulation in or cilitate both preclinical and clinical use for BBB opening and drug de- alongthe arteries (Fig.1c).This suggests that RaSP sequences livery to treat neurodegenerative disease and brain tumors. may reduce the likelihood and magnitude of unnecessary treat- ment of the arteries. Acoustic emissions from our short-pulses were more stable than those from the longpulses, with the en- ergy smoothly decreasing over time. The energy levels of the O30 acoustic emissions from the exposure to long-pulses varied Rapid short-pulse (Rasp) sequences – A new therapeutic greatly between pulses and within each pulse. ultrasound exposure paradigm to enhance drug delivery to the brain in vivo CONCLUSIONS These results suggest that RaSP sequences can im- Sophie Morse, Antonios Pouliopoulos, Tiffany Chan, Julien Lin, prove the spatial distribution of dextran delivery to the brain by pro- James J. Choi ducing a more uniform distribution of acoustic cavitation activity. Bioengineering, Imperial College London, London, UK Low pressure RaSP sequences could deliver a more efficient and safe Correspondence: Sophie Morse dose of drugs across the blood-brain barrier to treat diseases such as Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O30 Alzheimer’s, Parkinson’s and other neurodegenerative diseases. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 20 of 122 This abstract is not included as it has already been published: Nappoli A, Zaccagna F, Catocci G, Giulia B, Caliolo G, Andrani F, Catalano C. Magnetic resonance guided focused ultrasound surgery (MRgFUS) treatment of osteoid osteoma: a prospective development study. J Ther Ultrasound. 2015; 3(Suppl 1): O44. Available from: https://jtultrasound.bio- medcentral.com/articles/10.1186/2050-5736-3-S1-O44 O32 Transoesophageal HIFU for cardiac ablation: experiments on beating hearts 1 2 1 2 Paul Greillier , Bénédicte Ankou , Ali Zorgani , Francis Bessière , 3 3 4 4 Fabrice Marquet , Julie Magat , Sandrine Melot-Dusseau , Romain Lacoste , 3 5 2 1, 6 Bruno Quesson , Mathieu Pernot ,PhilippeChevalier , Cyril Lafon 1 2 LabTau - U1032, INSERM, LYON, Rhône, France; Hôpital Louis-Pradel, Lyon, 3 4 France; IHU-LIRYC - CHU Bordeaux, Pessac, France; Station de primatologie -CNRS- UPS846, Rousset, France; Institut Langevin - Ondes et Images - ESPCI ParisTech, CNRS UMR 7587, Paris, France; University of Virginia, Charlottesville, Virginia, USA Correspondence: Paul Greillier Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O32 Fig. 1 (abstract O30). In vivo brain drug delivery distributions using This abstract is not included as it has already been published: (a, b) rapid shortpulse (RaSP) or (c, d) conventional sequences. The Greillier P, Ankou B, Bessière, Zorgani A, Pioche M, Kwiecinski W, (a, c) left hippocampus of the mouse brain was sonicated in vivo Magat J, Melot-Dusseau S, Lacoste R, Quesson B, Pernot M, Catheline after systemic delivery of fluorescentlylabelled 3kDa dextran and S, Chevalier P, Lafon C. A75 Trans esophageal HIFU for cardiac abla- microbubbles. The (b, d) right hippocampus was not treated and tion: first experiment in non-human primate. J Ther Ultrasound. 2016; acted as our control (b, d). Sonication using (a) a RaSP sequence 4(Suppl 1):31. Available from: https://jtultrasound.biomedcentral.com/ delivered a greater dose of dextran to the brain parenchyma with articles/10.1186/s40349-016-0076-5 less delivery to arteries than (c) conventional sequences O33 In-vivo investigation of the combination of focused ultrasound and radiotherapy, using photoacoustic imaging as aplanning and monitoring tool Marcia M. Costa, Anant Shah, Ian Rivens, Tuathan O'Shea, Carol Box, Jeff Bamber, Gail ter Haar Radiotherapy and Imaging, The Institute of Cancer Research, Sutton, UK Correspondence: Marcia M. Costa Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O33 OBJECTIVES Tumour hypoxia is a limiting microenvironmental factor for radiotherapy (RT) success, since decreased oxygenation renders cells radioresistant, whereashigh intensity focused ultrasound (HIFU) tissue ablation is independent of oxygen levels. Imaging blood oxy- genation, which in part measures hypoxia, can be achieved non-inva- sively using photoacoustic imaging (PAI). The technique relies on the generationof acoustic waves (1-50MHz) by the tissue upon absorp- tion of short pulses of light. These waves can be detected using an ultrasound transducer and an image of relativeoptical absorption, Fig. 2 (abstract O30). Quantification of the normalised drug dose can then be reconstructed providing anatomical information. Further- and drug distribution delivered using RaSP and conventional more, light is absorbed differently by oxy- (HbO2) and deoxy-haemo- sequences. The normalised drug dose was quantified with the globin (Hb), and it is thus possible to calculate their relative normalised optical density (NOD), i.e. the average increase in proportions and, consequently, the distribution of oxygen saturation fluorescence in the targeted area normalised by the control. The (sO2=HbO2/(Hb+HbO2)) in tissue [1]. In this paper, we describe a drug distribution was quantified with the coefficient of variation combinatorial approach of HIFU and radiotherapy, using the former (COV), i.e. the standard deviation over the average fluorescence to target hypoxic (radioresistant) tumour volumes, in order to investi- intensity in the targeted region (* p < 0.01; ns = not significant) gateif this improves treatment outcome. PAI was used for HIFU treat- ment planning and assessing the treatment ouctcome. Furthermore, the relationship between the pretreatment sO2 values of tumours and radiotherapy response was also investigated. METHODS This study used a subcutaneously implanted human head O31 and neck tumour (CALR) in immunosuppressed mice, which was Magnetic resonance guided focused ultrasound surgery (MRgFUS) found to produce relativelyhypoxic tumours at a tumour size suitable Treatment of osteoid osteoma: a prospective development study for studies. Ultrasound imaging guided HIFU-alone (N=12) and com- Maayan Kimhy, Alessandro Napoli, Marco Marra Marcozzi puted tomography imaging guided radiotherapy-alone (N=13) treat- Dipartimento di Scienze Radiologiche, Oncologiche E Anatomo- ments were performed using dedicated small animal platforms: VIFU- Patologiche, Sapienza Università di Roma, Rome, Italy 2000 (Alpinion, Washington, USA) and SARRP (225KeV manufacturer Correspondence: Maayan Kimhy details, Xstrahl, Camberley, UK), respectively. Tumours had a mean Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O31 volume of 209 (+/-23 mm3). PAI (Multispectral Optoacoustic Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 21 of 122 Tomography (MSOT), iThera, Munich) was used to measure blood dis- tribution and its oxygen saturation non-invasively with the aim of identifying a main hypoxic tumour region for HIFU exposure. The HIFU treatment regime was a ‘spiral in’ pattern of 6 exposures, ISPTA -2 (free field) = 1.1 kW.cm and 10 second duration, per exposure. Exposures were done 1mm apart, in two rows, 3 exposures per row. Uniform whole tumour radiation exposures of 10 Gy, in a single fraction, were used. Animals were followed-up 60 days after treatment, using calliper measurements to calculate tumour volume. Once the treatment responses for individual treatments had been established, HIFU and RT were combined (total N=25), these being delivered 15 minutes apart. Both RTfollowed by HIFU and (N=12) the reverse treatment order (N=13) were investigated. RESULTS The HIFU-alone treatments resulted in 6 animals with no Fig. 2 (abstract O33). Treatment response A: proportion of tumour control and another 6 with an average tumour growth delay responders (R) and non-responders (NR) of 10gy-alone and combined of 6 (+/-4) days. The resultant hypointense lesion appearance in the treatments, showing an increase of responders for the latter group. B: Photoacoustic Imaging scan immediately after treatment is shown in Survival curves show an improvement in the overall survival of animals Fig. 1B. Surrounding the lesion, is a hyperintense region which spec- treated with the combined treatment compared to RT treatment alone. tral analysis showed to have increased haemoglobin content com- Arrows point to the average survival time of each treatment group pared to the pre-HIFU imaging (Fig. 1A). It is possible that this increase in haemoglobin signal is due to resultant haemorrhage (vas- cular disruption) caused by HIFU. Further histological studies are be- O34 ing undertaken to confirm these results. Radiotherapy alone Intracellular vaporization of antibody - conjugated phase - change treatments resulted in a 46% treatment response, defined as full nano-droplets for selective cancer therapy 1 1 2 1 tumour remission or growth inhibition within the follow up-period. Ayumu Ishijima , Takashi Azuma , Kosuke Minamihata , Satoshi Yamaguchi , 1 1 1 1 Spectral analysis of the photoacoustic signal showed a statistically Shinya Yamahira , Etsuko Kobayashi , Mariko Iijima , Yoshikazu Shibasaki , 1 1 significant difference between the sO2 values of responders (high TeruyukiNagamune ,IchiroSakuma 1 2 sO2) and non-responders (low sO2). The combined treatments out- The University of Tokyo, Tokyo, Japan; Kyushu University, Fukuoka, come, compared to the 10Gy-alone, showed an improvement in Japan treatment response both for percentage of responders (32% for RT Correspondence: Ayumu Ishijima +HFU and40% for HIFU+RT) and survival (Fig. 2). Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O34 CONCLUSIONS This preliminary study in a cell line known to be rela- tively radioresistant has shown a therapeutic benefit from combining OBJECTIVES Recent advances in nanomedicine provide the oppor- hypoxia targeted HIFU with whole tumour, in terms of improved out- tunity to reduce systemic toxicities of chemotherapies. However, come at lower doses. This effect is possibly due to the combination drug resistance remains a major challenge in cancer treatment re- of ablating radioresistant hypoxic areas and using RT for further kill search. Here we developed a nanomedicine composed of a phase- of the cells in better vascularised margins. Furthermore, PAI analysis change nano-droplet (PCND)1,2 and an anti-cancer antibody (9E5), showed the possibility of using this technique in adapting the treat- proposing the concept of ultrasound cancer therapy with intracellular ment planning forradiotherapy in clinic, as a relationship between vaporization to physically treat cancer. the sO2 measured values in tumour and the success of treatment METHODS 9E5-conjugated PCNDs (140 ± 120 nm) consists of a PFC was established. This project could have a major positive impact on liquid core (a mixture of perfluoropentane and perfluorohexane), a the treatment of tumours that are characteristically hypoxic, such as phospholipid shell, and antibody 9E5. The 9E5 human anti-epiregulin those of head and neck, if such combinationtreatments can be trans- (EREG) antibody was selected for active targeting of PCNDs. EREG, lated into clinical practice. [1] Xia et al., Electromagn Waves (Camb). the cell-membrane–expressed ligand of epidermal growth factor re- 2014; 147: 1–22. ceptor, is expressed and integrated into the plasma membrane at relatively high levels in a variety of human cancers.3 9E5 was conju- gated to PCNDs using the biotin streptavidin-biotin binding tech- nique. Biotinylated-9E5 and Alexa Fluor 647-conjugated streptavidin (SA-AF647) were bound to biotinylated PCND.The human colonic adenocarcinoma cell line DLD1 and the human gastric cancer cell line AGS were selected as high and low EREG-expressing cancer cell lines, respectively. To demonstrate the selective targeting capability of 9E5-conjugated PCNDs to DLD1 cells, we used SA-AF647 as a fluorescence probe. The targeted cells were observed by confocal laser scanning microscopy (CLSM), and the number of PCNDs at- tached and the fraction of bound DLD1 cells were quantitatively measuredby flow cytometer. Next, PCND-accumulated cells were ex- posed to ultrasound (100 cycles, 4 MHz, a peak negative pressure of 1.5 MPa), and its cytotoxic effects were visualized. The intracellular vaporization of 9E5-conjugated PCNDs in DLD1 cells was observed by a high-speed imaging system recording 101 subsequent frames at 0.25 Mfps. Furthermore, we quantitatively evaluated the cytotoxic Fig. 1 (abstract O33). A: Pre-HIFU photoacoustic imaging scan of R capabilities of vaporized PCNDs using PI staining and flow cytometer CAL tumour. B: Post-HIFU scan showing the lesion (red arrow) and after exposure of 5 cycles at 5 MHz of pulsed ultrasound with a peak a hypertensity rim (blue arrow) negative pressure of 4.6 MPa using an ultrasound imaging probe. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 22 of 122 RESULTS CLSM images of DLD1 cells treated with 9E5-conjugated OBJECTIVES The utility of Time Reversal Cavities (TRC) for medical PCNDs show the fluorescence signal inside the cells after 3 hours of ultrasound has been demonstrated for several applications. They incubation (Fig. 1a), whereas no clear fluorescence signals were ob- have indeed shown promising results for shock wave therapy, as served from the other types of cells. Flow cytometry analysis (N = 5) they allow high intensity pulses focusing in large regions of interest indicated that the 97.8 ± 0.5% of DLD1 cells were targeted by 9E5- with only electronical steering and a limited number of transducer el- conjugated PCNDs, whereas 1.4 ± 0.3% of DLD1 cells were targeted ements [1, 2]. TRCs have also been used as compact emitter- by non-9E5-conjugated PCNDs, similar to that of the control (4.4 ± receivers for pulse echo imaging [3 - 5]. In this work we present a 1.6%). Furthermore, the ratio ofAGS cells targeted by 9E5-conjugated dual mode cavity that can both perform shock wave therapy and 3D and non–9E5-conjugated PCNDs were close to the levels of the con- imaging in a large region of interest with 128 transducer elements. trol. The intracellular vaporization observed by high-speed imaging This dual mode capability is particularly interesting in transcostal ap- revealed that cell membranes were ruptured or broken into several plications where part of the therapeutic beam is shadowed by the parts during this initial stage ofvaporization (Fig. 1b). High-speed im- ribs lowering significantly the focal pressure. 3D volumetric imaging ages clearly show that intracellular vaporization caused a significant of the ribcage is used to perform an adaptive focusing of the thera- disturbance in cell morphology and destroyed the cells. The fraction peutic beam by transmitting selectively the ultrasonic energy of viable cells after ultrasound exposure was measured by flow cy- through the ribs. Adaptive focusing results in an increased efficiency tometry (N = 5). Cell viability was significantly reduced to 43.0 ± and higher pressure at the focal zone. 5.6% for 9E5-conjugatedPCNDs-treated DLD1 cells, while there was METHODS The leaky TRC is made of 2 orthogonal steel rod forests in no significant cell viability decrease for PCNDs without 9E5 conjuga- a reverberating cavity. The cavity itself is 20*5*5 cm with steel walls, tion (95.8 ± 2.0%) and without ultrasound exposure. Furthermore, and filled with water. A high power matrix array transducer (128- the viability of AGS cells did not decrease (98.0 ± 0.2%). These data elements, 1 MHz, Imasonics, France) is placed in the back of the indicate that PCND conjugated with 9E5 can sufficiently kill DLD1 cavity, opposite the aperture. For shock wave therapy experiments, cells with high selectivity. The addition of free 9E5 to DLD1 cells be- the probe was driven by custom multi-channel electronics (Corelec, fore treating/co-treating with 9E5-conjugated PCNDs significantly re- France). 40 μs US pulses emitted through the cavity were spread to duced the number of PI-stained cells (89.5 ±10.2%). up to 1 ms, picked up by a HNC 200 hydrophone (Onda, Sunnyvale, CONCLUSION In order to obtain cytotoxicity with droplet CA) and stored. This process will further on be called calibration step. vaporization, previous reports combined anti-cancer drugs such as Time reversal focusing (TRF) by compressing these signals in space doxorubicin with droplets.4 Our studies revealed that anticancer and time allowed us to reach the needed pressures. Steering the drug free 9E5-conjugated PCNDs are selectively internalized into tar- focal spot over a large volume was achieved by moving the hydro- geted cancer cells and kill the cells dynamically by ultrasound-in- phone (42*42 mm2 area, 1.5 mm grid step). We then re-emitted the duced intracellular vaporization. In vitro experiments show that 9E5- reversed signal at a pulse repetition frequency of 260 Hz, targeting conjugated PCND targets 97.8% of high EREG-expressing cancer cells an Ultracal phantom to form lesions. For imaging experiments, the and kills 57% of those targeted upon exposure to ultrasound. Fur- probe was driven by a Verasonic HIFU system. The same calibration thermore, direct observation of the intracellular vaporization process step was used, but instead of TRF, inverse filter (IF) focusing was revealed the significant morphological alterations of cells and the re- used. The focus quality was assessed by hydrophone scanning. The lease of intracellular contents. This work was supported by a grant signals corresponding to a focus on each grid point were succes- for the TSBMI from the MEXT of Japan. A.I. was supported by a sively emitted in free field, and the backscattered signals were re- Grant-in-Aid from the JSPS Research Fellows. corded on the transducers and stored, to constitute a reference library. An object was then placed in the focal plane, and the same References process was repeated. For each grid point, the reference was sub- 1. Kawabata K et al. Jpn J Appl Phys 2005; 44: 4548. tracted from the backscattered signals, and an image of the object 2. Ishijima A et al. Ultrasonics 2016; 69: 97–105. was reconstructed using the IF signals for focusing on reception. We 3. Lee YH et al. Biochem Biophys Res Commun 2013; 441: 1011–1017. used 2D simulations (Acel) to evaluate the pressure gain with select- 4. Wang CH et al. Biomaterials 2012; 33: 1939–1947. ive propagation of the therapeutic beam through the intercostal space. Calibration step on grid points in front of the cavity aperture was performed in free space. Reflectors mimicking a rib cage were then added in the propagating medium, blocking part of the aper- ture. The grid points were used as virtual transducers, and a delay law for a focus behind the ribs was applied. Either all the grid points or only those in the intercostal spaces were used. The total power was the same. Peak pressure was recorded on target in both cases. RESULTS Hydrophone measurements showed the device could cre- ate thin isotropic focal spots in a 200 cm region of interest, with very limited decrease of the peak pressure Therapy: Targeting the Ultracal® phantom in 7 different positions, we formed clear isotropic lesions in a 2.5x4 cm region, with only electronical steering. Imaging: The focal spots obtained with IF focusing showed central lobes with a 2 λ wide - 6 dB area, and a global -20 dB contrast. Side Fig. 1 (abstract O34). (a) Internalization of AF647labeled lobes were visible in one direction, but remained 10 dB below the 9E5conjugated PCND. CLSM observation of 9E5 antibodymediated main focus. We imaged either 2 steel wires, or 2 human ribs. In both accumulation and internalization of AF647labeled9E5conjugated cases, the object was clearly visible on the final image. The imaging PCNDs into DLD1 cells. (b) Highspeed imaging of intracellular vaporization window (4x4cm ) coincides with the device aperture. Coupling imaging and therapy: Simulations showed a 21 % increase in the focal peak pressure when turning off the virtual transducers placed O35 in front of the ribs Dual mode time reversal cavity for US shockwave therapy and 3D CONCLUSIONS We built a dual mode TRC that can both perform US imaging shock wave therapy and create a 3D image of the surrounding 1,2 1 1 1 J. Robin , B. Arnal , M. Tanter , M. Pernot medium. Though the image resolution is quite low due the low fre- 1 2 Institut Langevin, Paris, France; Université Paris 7, Paris, France quency used, it is sufficient to visualize strongly reflecting structures Correspondence: J. Robin like the rib cage. This could be particularly interesting in transcostal Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O35 therapy of the heart or the liver. Preliminary results indeed show that Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 23 of 122 selective propagation of the therapeutic beam through the intercos- O37 tal space would increase the peak pressure on target. Application of MR-ARFI to monitoring the stiffness changes of porcine muscle after HIFU therapy References Yangzi Qiao, Chao Zou, Xin Liu, Hairong Zheng [1] Arnal et al, Appl Phys Lett, 101 1-5, 2012 Shenzhen Institutes of Advanced Technology, Chinese Academy, [2] Robin et al, IEEE IUS, 2015 Shenzhen, China [3] Sarvazyan et al, Acoust Phys 55 630–7, 2009 Correspondence: Yangzi Qiao [4] Luong et al, Sci Rep 6 36096, 2016 Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O37 [5] Montaldo et al, Appl Phys Lett 2004 (No Image Selected) OBJECTIVES HIFU with the capability to treat deep tumor precisely is a fast developing therapeutic method. Due to the multiple contrast and unique thermometry ability, MRI has been the most popular im- O36 aging modality for HIFU guidance. Stiffness is an intrinsic property of The effects of steroids on the myocardial reduction induced by tissue, and can be monitored by MR-ARFI during HIFUtherapy. How- myocardial cavitation-enabled therapy (MCET) 1 2 2 2 2, 1 ever, the MR-ARFI based tissue stiffness evaluation results were quite Y. I. Zhu ,X.Lu , C. Dou , D. L. Miller , O. D. Kripfgans different in reports. It is expected that the tissue becomes stiffer after O.D. Kripfgans, Biomedical Engineering, University of Michigan, Ann coagulation, producing decreased displacement. But in some other Arbor, Michigan, USA; Department of Radiology, University of Michigan, reports, the coagulation or stiffer implant leads to an increased dis- Ann Arbor, Michigan, USA placement. The inconsistency of the reports may due to the different Correspondence: Y. I. Zhu acoustic parameters of different tissue. In this study, MR-ARFI was ap- Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O36 plied to monitoring the stiffness changes of porcine muscle during HIFU therapy. The results demonstrate that ARFI induced displace- OBJECTIVES Myocardial Cavitation Enabled Therapy (MCET) has been ment change is strongly correlated to coagulation. The displacement proposed as a means to achieve minimally invasive myocardial reduc- difference map can be used to depict the coagulation region, espe- tion using ultrasound to produce scattered microlesions by cavitating cially for small coagulation, making MR-ARFI an important comple- contrast agent microbubbles. Previous studies have demonstrated the mentary method for tissue monitoring during HIFU therapy. efficacy of MCET to induce damage. The purpose of this work was to METHODS All experiments were conducted on a 3T MR system (TIM study the treatment effect of the steroid methylprednisolone (MP), in Trio, Siemens). Ex vivo porcine muscle were sonicated by a 1.0MHz terms of swelling reduction following MCET and inhibition of fibrous tis- HIFU transducer (Imasonic). The input electric power was set from sue formation during long-term healing process. 8W to 100W, and transferred to the samples continuously in 10s to METHODS Dahl/SS rats from Charles River were used as an in vivo 60s. The segment SE-EPI (sEPI) sequence was usedfor ARFI imaging. model for HCM. Under ketamin/zylazine IP anesthesia, contrast agent The amplitude of displacement-encoding gradients (DEG) was 32mT/ was infused at a rate of 5μL/min/kg (tail vein catheter). The shaved and m, with input HIFU power of 55W and duty cycle of 2.5%. The dur- depilated left thorax was aimed at with a cardiac phased array (10S, ation of each DEG was 10ms. Vivid 7, GE Healthcare) to center on the left ventricular myocardium. In Imaging protocol was: TR=600ms, TE=36ms, slice thickness=5mm, this arrangement a 19 mm diameter single element therapy transducer resolution=1.6*1.6mm2, matrix=54*128, EPI factor=9. Acquisition was co-aligned to aim at a registered region of interest in the field of time=8.4s. Two sets of images with opposite polarity of DEG were view of the 10S array. For therapy 10-cycle tone bursts at 1.5 MHz, 4 used to quantify the displacement in each measurement. Ten mea- repetitions at 0.25 ms pulse interval, i.e. 4.0 kHz PRF, were sent every 8 surements of ARFI were scanned before and immediately after HIFU heartbeats, aligned with trigger end systole (RR/3, using ECG gating). sonication. T-test was used to determine whether tissue displace- Peak negative free field pressures of 4 MPa were used to induce cavita- ments have significantly changed. Temperature rise was monitored tion for 10 min. Therapy and sham therapy groups were followed up by GREduring HIFU sonication. The protocol was: TR=29ms, TE=10ms with MP administered at 0, 3, 6 and 24 hours after ultrasound exposure. with the same FOV and resolution. The ambient temperature was 19° Specifically, 30 mg/kg was chosen as high dose while 1 mg/kg was C. T2w image was acquired after HIFU sonication with TR=5000ms, used as low dose alternative. Myocardial wall thickness 24 hours after TE=89ms, resolution=0.8*0.8 mm2, matrix=108*256. therapy, measured from echocardiography was used to gauge the ef- RESULTS Figure 1(a) shows the displacement maps overlaid on ana- fect of initial myocardial swelling. White blood cell count was carried tomical image. No coagulation was found with 60 s sonication under out 24 hours after therapy. Hearts were removed after 4 weeks and ex- 32 W power. The peak temperature change was 17.3°C in the end of amined for evidence of the MP treatment effect. Histological sections the sonication, resulting in a peak temperature around 36°C. The av- with Masson’s trichrome staining were quantitatively analyzed for ex- eraged maximal displacement before sonication was 3.41±0.46 μm, tent of fibrosis, i.e. tissue scarring. while the averaged maximal displacement after sonication was 3.36 RESULTS Myocardial wall thickness from echocardiography was mea- ±0.26 μm. The maximal displacement was constant during the suc- sured as: sham 1.69±1.6 mm; therapy only 2.68±0.21 mm; therapy with cessive 10 measurements after sonication, indicating the maximal low MP 2.29±0.22 mm; and therapy with high MP 2.01±0.14 mm. High displacement might be independent of temperature change (Figure dose of MP had a 25% wall swelling reduction with a p-value of 0.003 1(b)). It is also shown in Figure 1(c) there is no signification difference relative to the therapy only group. Absolute neutrophil count demon- between the maximal displacements before and after HIFU sonic- strated the efficacy of MP: sham 5.0±1.1 109/L; therapy only 5.4±0.9 ation with p = 0.18 from t-test. Figure 2 shows another case with in- 109/L; therapy with low MP 3.6±1.4 109/L; and therapy with high MP put power 82 W under 30s sonication, coagulation occurred. The 2.7±1.2 109/L. High dose of MP had a 45% neutrophil count reduction peak temperature was around 85°C (peak temperature change was though with a p-value of 0.026 relative to the therapy only group. Fi- 66.7°C), much higher than the protein denature temperature. The dis- brosis densities were quantified for the treated regions as to represent placement maps before and after sonication differed a lot. The max- the extent of scarring with results therapy only 25.6±4.0%; therapy with imal displacement significantly increased after sonication, with an low MP 25.7±1.6%; and therapy with high MP 20.6±0.4%. High dose averaged increment of 1.86 μm (p<0.01). The averaged maximal dis- MP on average reduced fibrosis formation of 20% though is underpow- placement before sonication was 3.46±0.17 μm, while the averaged ered in this study with a p-value of 0.154. maximal displacement after sonication was 5.32±0.28 μm. T2w image CONCLUSIONS In this MCET rodent study, reduction trend of swelling and displacement difference map (the difference between the dis- wall thickness and neutrophil count has shown that MP is effective to placement before and after HIFU sonication) were compared for inhibit the inflammatory response and reduce swelling right after ther- visualization of thecoagulation. In T2w image, the coagulation was apy. Additionally, long-term study of fibrosis density analysis indicated pointed out by a red dashed circle (Fig. 3). In displacement difference that MP potentially helps reducing fibrosis formation. Cavitation en- map, the region where displacement before and after HIFUsonication hanced therapy maybe a possible avenue for non-invasive tissue reduc- show significant difference (p<0.01) was overlaid on the anatomic tion for treatment of hypertrophic cardiomyopathy. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 24 of 122 image. In another sonication (with input power 82W, 10s sonication), The peak temperature was around 48.2°C (peak temperature change was 29.2°C). However, the coagulation can be hardly recognized in T2w image, while it is still visible in displacement difference map. The averaged maximal displacement before sonication was 3.45±0.12 μm. While the averaged maximal displacement after sonication was 5.06±0.19 μm. The coagulation area is marked in Fig. 4(c). CONCLUSIONS In conclusion, MR-ARFI can detect significant changes of porcine muscle displacement while coagulation occurred. After heating the displacement significantly increased at the coagulation region. The displacement difference map can be used to visualize and evaluate therapy effect, especially for small coagulation. This makes MR-ARFI an important complementary method for tissue monitoring during HIFU therapy. Fig. 3 (abstract O37). (a) T2w image, (b) displacement difference map, and (c) photo of coagulation with input power of 82W, 30s continuous sonication Fig. 1 (abstract O37). (a) The displacement map before (left) and after (right) HIFU sonication. (b) The maximal displacement in focus. (c) The averaged maximal displacement with inputpower of 32W, 60s continuous Fig. 4 (abstract O37). (a) T2w image, (b) displacement difference map, and (c) photo of coagulation with input power of 82W, 10s continuous sonication O39 Long-term oncologic outcomes of focal magnetic resonance guided focused ultrasound treatment for locally confined prostate cancer Alexander Nosov, Sergey Reva, Maria Berkut Fig. 2 (abstract O37). (a) The displacement map before (left) and Onco-urological Department, N.N. Petrov Research Institute of Oncology, after (right) HIFU sonication. (b) maximal displacement in focus. (c) Saint-Petersburg, Russian Federation The averaged maximal displacement with input power of 82W, 30s Correspondence: Alexander Nosov continuous sonication Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O39 Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 25 of 122 OBJECTIVES Progress in different diagnostic and treatment mo- tremor score, and T2-weighted magnetic resonance imaging (T2- dalities of prostate cancer (PCa) have strengthened support for MRI). Each of these methods provides an imperfect estimate of the the use of focal high-intensity focused ultrasound (HIFU). How- ultimate treatment effect. Diffusion-weighted MR imaging (DW-MRI) ever, important questions remain regarding candidate selection, can potentially provide accurate lesion visualization and prediction at treatment, and outcomes. We assessed long-term oncologic out- earlier time points than T2-MRI. This is analogous toischemic stroke, comes of focalHIFU in a small single-center cohort of low-risk where DW-MRI is able to observe infarcts at early time points. To PCa patients. date, DW-MRI has achieved very limited success because of technical METHODS Twenty-two patients with low risk PCa (PSA<10 ng/ difficulties introduced by the focused ultrasound system such as ml, Gleason score less than 7, or clinical stage cT2b and less) eddy currents, magnetic field inhomogeneity, and patient motion. A were underwent for focal HIFU ablation (ExAblate 2100 for a dual-echo, steady-state sequence has found moderate success at prostate device, InSightec) with GE MRI suite and endorectal identifying lesions in a pig model [1]. Here, we propose a single-shot, FUS transducer from March 2009 to January 2010. Among them, spiral DW-MRI sequence coupled with an outer volume suppression 8 patients were available for long term (a median time – 7.3 preparation sequence [2] as a means to quickly obtain diffusion years) follow-up. Desired treatment target (Region of Treatment, weighted images of sub thalamic nuclei. See Fig. 1. The single-shot, ROT) and vulnerable structures (nerve vascular bundles) were spiral acquisition reduces sequence sensitivity to bulk motion and marked on acquired magnetic resonance guided (MRg) planning eddy currents. Meanwhile, the outer volume suppression pulse re- images. Pre- and post-reatment strategy, rate and follow-up duces the imaging field of view to a region that can be successfully schedule was designed at PCa001 andPCa002 studies and de- supported by the spiral trajectory. scribed previously; prospective parts of these studies was closed METHODS The proposed sequence is displayed in Fig. 1. A single- after 6-month follow up. shot, spiral acquisition is used during the readout portion of a Stejs- RESULTS The average patient’s age 64 (49-73) years. Median pre- kal-Tanner diffusion-weightedsequence. The spiral gradient samples HIFU PSA level and post-HIFU PSA nadir was 7.6 and 3.9 ng/ml, a 4 cm diameter circular plane at 1.5 mm resolution using a variable respectively. Biochemicalrecurrence (BCR, defined as nadir + 2 density trajectory of 10 ms duration. The sequence is prepended by ng/ml) was observed in 7 (87.5%) cases. Medium time to progres- an outer volume suppression pulse consisting of an 8 ms long BIR-4 sion was 18 (3-32) months. In 4 (50%) cases local progression adiabatic excitation pulse and a 4 ms long cylindrical tip-back pulse was confirmed by prostate biopsy and after metastatic process as described in [2]. The preparation pulse reduces the field of avail- exclusion salvage radical prostatectomy (RPE). Generally, surgery able magnetizaiton to a cylinder approximately 3.5 cm in diameter. after ablation was severe than in naïve patients; however, opera- This sequence was validated using the the bodycoil of a 3T MR scan- tive characteristics (operative time, blood loss, hospital stay) were ner (GE, Waukesha, WI) and a human volunteer. Imaging parameters comparable with historical cohort. During follow-up time, sys- are; TR: 1 s, TE: 45 ms, FOV: 5 cm, resolution: 1.5 mm, slice thickness: temic treatment (hormonal therapy) was prescribed for 5 patients, 0.8cm, bValue: 750 smm-2, averages: 16. We will measure the time as a result of distant metastatic progression after prostatectomy course of the diffusion-weighted signal after sonication by subjecting (2) and without secondary local treatment (3). Cancerspecific sur- two instances of a porcine model with craniotomy to focused ultra- vival (CSS) and overall survival (OS) was both 100%. soundablation (ExAblate Neuro, INSIGHTEC, Haifa, Israel) while under CONCLUSIONS Despiteacceptableresults of survival,wefound anesthesia. A total of 8 lesions will be made in each animal and DW that almost all patients are progressed during follow-up. These and T2-weighted images will be acquired at time points spaced by 2 data are in contrast with previously published data. However, minutes from immediately after sonication to 20 minutes after sonic- patients in our series were younger than in historical cohort. ation using the proposed sequence for DW-MRI and a fast-spin echo We concluded that the use of focal high-intensity focused ultra- sequence with spiral acquisitons for T2-MRI. We will then compare soundinselectedpatientsrepresents a strategy combining the time course of the DW-MRI signal immediately after ablation to benefit of active surveillance and radical treatment in patients the time course of the T2-MRI signal. with low risk PCa. However, this concept should be evaluated in RESULTS Figure 2A and B demonstrate the outer volume suppression large prospective controlled studies. sequence using an enlarged field of view and a six-shot, interleaved spiral acquisition. By applying the suppression preparation pulse, the field of view is successfully reduced to a region about the right ven- tricle. A single-shot spiral image with reduced field of view of this re- gion, as well as an identical image with diffusion weighting, are O40 displayed in Fig. 2C and D, respectively. Diffusion-weighted contrast Rapid diffusion-weighted imaging immediately after sonication can be observed in the suppressed signal of the ventricle. The poor using outer volume suppression pulses and single-shot, spiral transmit and recieve gains of the body coil do not prevent the suc- acquisition 1 1 1 2 cessful implementation of this sequence. Steven P. Allen , Xue Feng , Helen L. Sporkin , Jeff Elias , 3 1 CONCLUSIONS Single-shot, spiral acquisition with an outer volume Kim Butts-Pauly , Craig H. Meyer suppression preparation pulse shows promise as a motion-insensi- Biomedical Engineering, University of Virginia, Charlottesville, Virginia, tive sequence to capture diffusion weighted contrast in the USA; Neurological Surgery, University of Virginia Hospital, Charlottesville, thalamus. Virginia, USA; Radiology, Stanford University, Palo Alto, Virginia, USA Correspondence: Steven P. Allen Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O40 References [1] Plata et al. “A feasibility study on monitoring the evolution of apparent OBJECTIVES There remains a critical need to predict the long term diffusion coefficient decrease during thermal ablation,” Med. Phys., 42(8), effects of a sonication during MR-guided, focused ultrasound thala- 5130–5137, 2015 motomy. Current prediction methods include calculations based on [2] Smith et al. “Reduced field of view MRI with rapid, B1-robust outer the observed temperature trajectory, observations of the patient’s volume suppression,” Magn. Reson. Med., 67(5), 1316–1323, 2012. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 26 of 122 O41 Selection of MR-HIFU hyperthermia treatment sites based on MR thermometry evaluation in healthy volunteers 1 2 3 1 Satya V.V.N. Kothapalli , Michael Altman , Ari Partanen , Lifei Zhu , 1 2 2 2 Galen Cheng , H. Michael Gach , William Straube , Dennis Hallahan , 1,2 Hong Chen Biomedical Engieering, Washington University in Saint Louis, Saint Louis, Missouri, USA; Department of Radiation Oncology, Washington University in St. Louis, Saint Louis, Missouri, USA; Clinical Science MR Therapy, Philips, Andover, Massachusetts, USA Correspondence: Satya V.V.N. Kothapalli Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O41 OBJECTIVES Magnetic resonance imaging-guided high-intensity fo- cused ultrasound (MR-HIFU) is a promising noninvasive technique for mild hyperthermia, enabling targeted and localized hyperthermia Fig. 1 (abstract O40). The proposed diffusionweighted sequence. therapy deep within the body under real-time MRI-based An outer volume suppression pulse is prepended to a StejskalTanner temperature monitoring and control. However, the narrow thera- diffusionweighted imaging pulse using a singleshot spiral trajectory peutic window (41-43°C) and long treatment duration (30-60 mi- for image acquisiton nutes) pose a great challenge on MR thermometry. As a first step toward effective clinical translation of MRHIFU hyperthermia, this study identified preferable treatment sites based on the accuracy and precision of MR thermometry performed on healthy volunteers. METHODS A total of 15 healthy volunteers (age 18-45 y and body weight 45-90 kg) were recruited. A clinical MR-HIFU system (Sonalleve V2, Philips, Vantaa, Finland) was used with the patient tabletop docked into the MRI bore (Ingenia 1.5T, Philips, Best, the Netherlands). Volun- teers were positioned above the tabletop’sacousticwindow and a standard HIFU-compatible 3-ch pelvis RF receive coil was secured over the target anatomy. For MR image acquisition, the pelvis coil was used together with the 2-ch RF receive coil integrated in the HIFU tabletop. A dynamic multi-slice fast-field-echo pulse sequence (sequence (TR=41 ms; TE=19 ms;voxel=2.5×2.5×7 mm2; FOV=400×400 mm2)) was used for real-time MRI (without HIFU sonication) together with the proton reson- ance frequency shift (PRFS) method to calculate temperature maps. Eight volunteers were subjected to a shorter scanning period (5 mins) targeting the upper body, pelvis, and lower extremity. Seven volunteers were subjected to a longer scanning period (30 mins) targeting the lower extremities, i.e., thigh and calf. The precision of MR thermometry was quantified by the temporal temperature uncertainty, which calcu- lated the temporal standard deviation for each pixel within an 18×18 mm2 ROI. The accuracy of MR thermometry wasquantified within the same ROI by the absolute error of measured temperatures compared to body temperature (37°C). Temperature monitoring with both uncertainty and absolute error lower than 1°C is expected to be satisfactory for hyperthermia. RESULTS MR thermometry measurements based on 5-min scans at the chest, pelvis, and lower extremity had an uncertainty of 2.53°C ±0.48°C, 1.89°C±0.50°C, and0.50°C±0.04°C, respectively, and an abso- lute error of 0.63°C±0.63°C, 2.88°C±0.87°C, and 0.08°C±0.13°C, re- spectively. MR thermometry measurements based on 30-min scans Fig. 2 (abstract O40). Validation of the proposed sequence in at the lower extremity found the uncertainty and the absolute error a human volunteer using the body coil of a 3T MR scanner. of the MR thermometry to be 0.52°C±0.13°C and 0.12°C±0.06°C, re- (A) Image acquired with a nominal field of view. (B) The spectively. No significant difference was found between 5-min scan- same sequence as (A), but prepended with the outer volume ning and 30-min scanning for the lower extremity. Among the three suppression pulse. The preparation pulse reduces the region of anatomical sites, only the lower extremity had satisfactory excited magnetization to a small cylinder. (C) The reduced field temperature accuracy and precision according to the suggested of view imaged with a singleshot spiral acquisiton. (D) Same as criterion. (C) but with a diffusionweighting of 750 smm2. Diffusionweighting can CONCLUSIONS This study constitutes the first evaluation of MR be observed inthe suppression of the ventricle. The poor transmit and thermometry performance at different body sites for long scan recieve gains of the body coil do not prevent the successful times that are relevant in clinical MRgHIFU hyperthermia ther- implementation of this sequence apy. Motion induced by respiration and cardiac activity poses a Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 27 of 122 technical challenge on the application of this treatment on chest and pelvis. Theextremity was found to be a preferable site for MR-HIFU hyperthermia using the suggested criteria that both the uncertainty and absolute error need to be under 1°C. O42 Behaviour of bubbles generated by acoustic droplet vaporization within inter-tissue region Y. Ho, C. Yeh Department of Biomedical Engineering and Environmental Sciences, National Tsing Hua University Correspondence: Y. Ho Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O42 OBJECTIVES Treatment resistance is a critical consideration in cancer research. The tumor cells which extensively proliferate away from vessels show treatment resistance due to the restricted drug penetration and biological mutation of hypoxia. Acoustic droplet vaporization (ADV) can convert nanodroplets into gaseous bubbles (ADV-Bs) and simultaneously disrupt vessels to improves the penetration of nanodroplets, ADV-Bs, and drugs. Previous studies have been proved the mechanical Fig. 1 (abstract O42). See text for description force of bubble cavitation can directly induce in vitro and intra- vascular bioeffects. Therefore, our study investigated the behav- iors of ADV-Bs in vessels and tissue triggered by ultrasound O43 (US), and then evaluated the feasibility of using intertissue ADV- Enhanced ultrasound-mediated trans-membrane drug delivery into Bs to treat resistant tumor cells by physical damage. a monolayer of endothelial cells during high flow in vitro METHODS The window chamber model was used to directly observe conditions vessels and tissue pattern on mouse dorsal skin by the integrated Syed Mohd Nooh Syed Omar, Rob Krams, James J. Choi acousto-optical system. Mice were injected 30 μL NDs (304±97 nm) Bioengineering, Imperial College London, London, UK with the concentration of 1.2-1.4×10 NDs/mL to form ADV-Bs by 3- Correspondence: Syed Mohd Nooh Syed Omar cycle single-pulsed US at 10 MPa. The modulated parameters of US Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O43 with pulse repeat frequency of 1 Hz were used to stimulate ADV-Bs as following: (1) 100-cycle with 5 and 10 MPa for movement; (2) 3- OBJECTIVES One of the challenges in the treatment of atherosclerosis cycle with 5 MPa for cavitation. Furthermore, the transgenic adeno- and other cardiovascular disease is the inability to deliver therapeutic carcinoma of the mouse prostate (TRAMP) cells with live-cell imper- drugs across the cell membrane effectively and safely. Entire classes of meable dye propidium iodide (PI) were used to evaluate cellular drugs, nucleic acids (e.g., DNA, siRNA and mRNA) and biomolecules for im- bioeffect induced by ADV-Bs. proving treatment, are ineffective, because they cannot enter the cell at RESULTS Intravital images showed vascular disruption, intravas- safe systemic doses that do not cause side effects throughout the rest of cular ADV-Bs, and intertissue ADV-Bs after ADV. The intravascu- the body. Over the last several decades, ultrasound-mediated trans-mem- lar ADV-Bs were moved following the vascular shape and stuck brane drug delivery methods (e.g. sonoporation) have been studied at the branch (black arrows) during US stimulation (Fig. 1. A). in vitro to deliver drugs into cells. Through these studies, it was discovered On the other hand, the movement of intertissue ADV-Bs was that several mechanisms of trans-membrane drug delivery exist and that not restricted and the distance per pulse was 48±17 μmfor 5 they are highly dependent on the acoustic parameters, microbubble con- MPa and 88±26 μm for 10 MPa. Cavitation of intertissue ADV-Bs ditions, and the cell-type used. Despite promising results from these study, was observed to split into two bubbles and coalesce back to the advancement from single cell-bubble interactions to clinical use has one during US stimulation (Fig. 1. B). In Fig. 1. C, the intracellu- not been made. This gap in development is largely because the under- lar enhancement of PI indicated cell membrane damage after lying mechanism of trans-membrane drug delivery under high flow condi- ADV-Bs shoving (yellow arrows). The cell and ADV-Bs were tion and for a large population of cells, remains poorly understood and blown out of the original position at 15 s because the cell stuck poorly controlled. Our study explores the ultrasound and microbubble-me- on theADV-Bsand waspushedby pressure gradient of US. diated trans-membrane drug delivery efficiency and safety to a monolayer CONCLUSIONS This study revealed intravital imaging to observe ADV- of endothelial cells using a state-of-theart physiologically-relevant cultiva- B formation, cavitation, and movement in vessels and tissue. Intertissue tion system under different ultrasound exposure conditions. In the end, ADV-Bs could be pushed to distant tissue by US stimulation, and pro- we will evaluate the critical question on whether safe transmembrane duced cell membrane damage when ADV-Bs cavitated or shoved cells drug delivery can be achieved in such a complex, physiologically relevant during movement. Since the treatment resistant tumor cells live away environment. from vessels, these results provided a possibility to touch these cells METHODS Human umbilical vein endothelial cells (HUVEC) were se- and damaged them by ADV-B movement and cavitation. Moreover, the lected as our model cells since they are well described due to their ex- cellular bioeffect induced by physical damage also presented a poten- tensive use in vascular biology and are similar to the arterial endothelial tial way to overcome the resistance of drugs and radiation in tumor cells we’d like to treat in future studies (e.g., in Atherosclerosis). They therapy. were cultured as a monolayer in a flow chamberunder static or pulsatile Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 28 of 122 shear stress (10 dyne/cm2) and at 37°C for 3 days (Fig. 1a, b). A focused ultrasound transducer (0.5 MHz) sonicated the flow chamber in the pres- ence of lipid-shelled microbubbles and during flow conditions (fluid flow rate: 17 ml/min) (Fig. 1c). A centrally-inserted transducer (7.5 MHz) was used to determine the type of microbubble activity generated by pas- sively detecting cavitation emissions during sonication. After sonication, trans-membrane delivery was assessed by quantifying the intracellular uptake of propidium iodide (PI), which is a DNA-binding fluorescent dye Fig. 2 (abstract O43). The effect of acoustic pressure and pulse that is normally impermeable to the cell membrane. Cell viabilitywas duration on (a) Transmembrane delivery and (b) cell viability. The assessed by Calcein-AM. A range of ultrasound parameters (pressure, percentages of transmembrane delivery and cellviability are shown pulse length and exposure duration) and microbubbles conditions were with respect to pressure at 50, 75, 100 and 150 kPa with pulse length systematically evaluated to identify different trans-membrane delivery of 500 and 1000 cycles. Exposure condition: fc=0.5 MHz, PRF=1.4 Hz, and safety profiles. Np =500with microbubbles at 1X clinical dose. The results shown RESULTS The main finding of our study is that safe trans-membrane represents the mean for at least three independent measurements drug delivery can be produced in a clinically targetable cell type (i.e., endothelial cells) underphysiologically relevant conditions (i.e., high fluid velocities) (Fig. 1d). This safe delivery is indicated by cells where red propidium iodide is co-localised with green calcein. A range of O44 ultrasound parameters was explored and we were able to character- Enhanced angiogenesis and perfusion via delivery of basic ise different biological effects: cell viability with and without trans- fibroblast growth factor using an acoustically responsive scaffold 1,2 2 2 2, 1 membrane delivery, cell death and cell detachment. Trans- Alexander Moncion , Melissa Lin , Eric O'Neill , Oliver D. Kripfgans , 3, 4 4 1, 2 membrane delivery increased and cell viability decreased with pres- Renny T. Franceschi , Andrew J. Putnam , Mario L. Fabiilli sure ranging from 50 kPa to 150 kPa, at pulse length of 500 and Applied Physics Program, University of Michigan, Ann Arbor, Michigan, 1000 cycles and in the presence of microbubbles (Fig. 2). It was seen USA; Department of Radiology, University of Michigan, Ann Arbor, that at the highest pressure and pulse length evaluated, transmem- Michigan, USA; School of Dentistry, University of Michigan, Ann Arbor, brane delivery occurred in9.68% ± 10.96% of cells while only 13.18 ± Michigan, USA; Department of Biomedical Engineering, University of 7.33% maintained cell viability. At the lowest acoustic pressure and Michigan, AnnArbor, Michigan, USA pulse length evaluated, trans-membrane delivery occurred in 1.12% Correspondence: Alexander Moncion ± 1.35% of cells while cell viability was 92.45% ± 9.04%. There were Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O44 rapid transitions in cell viability and drug delivery efficiency between 75 kPa and 100 kPa. OBJECTIVES Regenerative growth factors (GFs) are expressed in a CONCLUSIONS This study demonstrated that safe trans-membrane spatially- and temporally-regulated manner during wound healing. drug delivery can be produced in a clinically targetable cell type (i.e., However, delivery of GFs using conventional hydrogel scaffolds endothelial cells) and under physiologically relevant conditions (i.e., does not afford such control, which has hindered the translation high shear stress condition). However, the trans-membrane delivery of GF-based therapies in tissue regeneration. We have developed efficiency was low. This implies that there is a specific microbubble acoustically-responsive scaffolds (ARSs), which are hydrogels dynamic within the exposed tissue volume that can produce the de- doped with sonosensitive perfluorocarbon emulsions containing sired bio-effect, but that it is being produced in only a small subpop- encapsulated GFs. When ARSs are exposed to ultrasound (US) ulation of microbubbles. In future works we will identify this safe and above the threshold for acoustic droplet vaporization (ADV), the effective dynamic and develop pulse sequencing technologies to GF is released from the emulsion. This work studies how ARSs control it so that we can maximise safe drug delivery. can be used to enhance angiogenesis in an in vivo model via the controlled delivery of basic fibroblast growth factor (bFGF), which can stimulate blood vessel formation. METHODS bFGF was encapsulated in monodispersed, perfluoro- hexane (C6F14) double emulsions (mean diameter: 13.4±0.04 μm). ARSs (0.3 mL volume, 10 mg/mLfibrin, 1% (v/v) emulsion) were prepared with 1 μg of bFGF per implant, and injected subcutaneously in the lower back of BALB/c mice. After polymerization, a subset of the implants were exposed to US (2.5 MHz, Pressure = 8 MPa peak negative, 13 cycles, 100 Hz pulse repetition frequency) for 2 min daily. To assess perfusion in and around the implants, the mice were imaged on days 0,1 3, 7, 10, and 14 with a PeriMed Laser Speckle Contrast Imaging (LASCA) system. Blood vessel density in the implants was de- termined via CD31 immunohistochemical staining on days 7 and 14. RESULTS LASCA (Fig. 1): A greater percent change in perfusion, rela- tive to day 0, was observed in ARSs exposed to US (i.e., +US) versus ARSs not exposed to US (i.e., –US) on days 7 (97.0 ± 14.9% vs. 54.0 ± Fig. 1 (abstract O43). Transmembrane drug delivery under flow 13.1%) and 10 (46.9 ± 7.0 vs. 14.3 ± 5.0), respectively. CD31 (Fig. 2): condition. Human umbilical vein endothelial cell (HUVEC) were stained for More blood vessels were present in the +US ARSthan the –US ARS cell viability by calceinAM (green cells) and transmembrane delivery by on day 7 (126.8 ± 23.8 vs. 73.1 ± 21.2 BVs/mm2). Sustained angio- internalization of propidium iodide (red cells). (a) Microbubbles were made genesis was observed with ARSs when compared to fibrin loaded to flow (fluid velocity: 95 mm/s) through a collagentreatedflow chamber with bFGF. with a channel height of 600 μm and a volume of 150 μL. (b) The regioin CONCLUSIONS US enables non-invasive, controlled release of (dotted rectangle) acted as nonsonicated control region, while (c) central bFGF from an ARS, which locally enhanced angiogenesis and per- chamber region acted as sonicated region (solid rectangle). (d) Zoomed in fusion. These finding could be applied in therapeutic angiogen- view at delivery region; a population of cell that exhibited a mixture of esis for the treatment of cardiovascular diseases. Future work will viable cells (green cells) with and without transmembrane delivery and dead attempt to demonstrate efficacy in ischemic pathologies as well cell (red cell) as controlled release of multiple bioactive molecules. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 29 of 122 transducer (3.57 MHz, 0.46x3.55 mm focal area) was used for stimulation while initially exploring the following parametric space: 0.7-5.4 MPa peak negative pressure, 4ms-1s stimulation duration,5-100% dutycycle,0.01- 1kHz PRF.Targetingof the nerve was conducted through a 18.5 MHz, 128-element imaging probe. Once efficacious parameters had been determined (the ability to elicit subsequent EMG responses after an initial re- sponse), the safety of the procedure was explored along with a comparison to electrical stimulation. H&E histology of the tar- geted region (n = 4 stimulation, n = 1 positive control, n = 1 negative control) were employed in a blinded study to detect damage (red blood cell extravasation, inflammation, cell mem- brane rupture). Mice (n = 4 stimulation, n = 4 control) com- pleted an open field behavioral test to explore the short-term safety of the experiment. FUS EMG responses (peak-to-peak, delay to signal) were compared to direct electrical stimulation (1-10 V, 1-10 mA, 200500μs). To measure temperature increase during FUS stimulation, thermocouples were embedded in Fig. 1 (abstract O44). Percent change in perfusion relative to day 0. ex vivo mouse limbs alongside the sciatic nerve and exposed to Peak perfusion was observed on day 7, while the +US condition had FUS stimulation. A force balance was used to determine the statistically greater perfusion both day 7 and 10 acoustic radiation force generated by the transducer to estimate the tissue deformation in the focal region. To confirm neural ac- tivity at the single-unit level, a ex vivo skin-saphenous nerve prep (n = 15) was also stimulated by FUS with the aforemen- tioned parameters while recording extracellular electrophysi- ology responses. RESULTS Non-invasive stimulation of the sciatic nerve via FUS was shown capable of eliciting both observable leg twitching and measurable EMG responses when using the following FUS param- eters: 0.2-5.7 MPa, 35-100% DC (continuous wave), 0.1-1kHz PRF, 0.8- 10.5 ms stimulation duration. Increasing the pressure and DC raised the average peak-to-peak EMG response along with the success rate (Fig. 1). Varying the PRF or stimulation duration did not have a significant ef- fect on response. Both delay and peak-to-peak EMG responses for FUS stimulation were found to be comparable to direct electrical stimula- tion of the sciatic nerve. Mice stimulated with efficacious parameters did not display any significant deviation in behavior compared to the control group or baseline values. The blinded histology study did not detect any damage for the stimulated group, only for the positive con- trol. In ex vivo experiments, FUS was able to stimulate action potentials Fig. 2 (abstract O44). Blood vessel counts determined using CD31 firing in multiple classes of peripheral neurons. Latency to action po- immunohistochemistry. Greater blood vessel formation was observed tential firing was dependent on sensory neuron type and FUS for ARSs with GF exposed to US on day 7 intensity. CONCLUSIONS FUS stimulation of the sciatic nerve was shown cap- able of eliciting physiological responses in vivo. This demonstrates that FUS can be a non-invasive alternative to conventional thera- peutic methods. Specific FUS parameters has been identified for suc- cessful and safe stimulation. Future work to explore the potential mechanisms of generation of the action potential will dictate the O45 FUS parameters to translate this technique to clinical applications. Non-invasive peripheral nerve stimulation via focused ultrasound M. Downs, S.A. Lee, Y. Baba, B. Hoffman, G. Yang, D. Kim, E. Lumpkin, E. Konofagou Columbia University, New York, New York, United States Correspondence: M. Downs Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O45 OBJECTIVES Focused ultrasound (FUS) has been demonstrated to modulate neuronal activity in both the central (CNS) and per- ipheral (PNS) nervous system. While modulation of the CNS has successfully elicited both motor and sensory physiological re- sponses, PNS modulation has only been shown to induce inhibi- tory effects, primarily ex vivo. The ability to non-invasively stimulate and inhibit the PNS with FUS would allow clinicians an alternative therapeutic modality to treat peripheral neur- opathy, as current treatment procedures can generate systemic side effects or require invasive procedures. In this study, we Fig. 1 (abstract O45). EMG responses to FUS stimulation of the aim to show that FUS can elicit excitatory effects targeting the sciatic nerve. The green and red vertical lines are the start and end PNS and generate downstream physiological responses. of FUS stimulation respectively. The blue line is the EMG response to METHODS Focused ultrasound was used to target the sciatic FUS stimulation. Stimulation parameters were as follows: 3.5 MHz, 3.7 nerve in mice while recording EMG responses from the tibialis MPa, continuous wave, 0.88ms stimulation duration anterior muscle in sedated mice (n = 25). A single-element HIFU Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 30 of 122 O46 O48 The safety and feasibility of high intensity focused ultrasound in Clinical experience of intra-operative high intensity focused treatment of resistant hypertension ultrasound in patients with colorectal liver metastases: Results of a P. You Phase II study 1,2 1, 2 2 2 1, 1, 2 State Key Laboratory of Ultrasound Engineering in Medicine Co-founded D. Melodelima ,A.Dupre , Y. Chen , D. Perol , J. Vincenot M. Rivoire 1 2 by Chongqing and the Ministry of Science and Technology,Chongqing Key LabTAU - U1032, INSERM, Lyon, Rhône Alpes, France; Centre Leon Laboratory of Ultrasound in Medicine and Engineering,College of Berard, Lyon, France Biomedical Engineering,Chongqing Medical University, Chongqing, China Correspondence: D. Melodelima Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O46 Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O48 OBJECTIVES To evaluate the safety and feasibility of high intensity This abstract is not included as it has already been published: focused ultrasound (HIFU) in the treatment of resistant hypertension. Melodelima D, Dupre A, Vincenot J, Chen Y, Perol D, Rivoire M. A49 Clin- METHODS 40 patients with resistant hypertension underwent the ical experience of intra-operative High Intensity Focused Ultrasound in treatment of high intensity focused ultrasound, Intraoperative and patients with colorectal liver metastases. Results of a Phase II study. J Ther postoperative adverse effects were recorded; the drop of blood pres- Ultrasound. 2016; 4(Suppl 1):31. Available from: https://jtultra- sure, types of drug used, peak systolic velocity of renal artery and sound.biomedcentral.com/articles/10.1186/s40349-016-0076-5. renal function was followed up to 6 months post treatment. RESULTS (1) All patients completed HIFU treatment successfully.The major discomforts that patients complained during the procedure was O49 pain in the treatment area,which usually disappeared within 24 hours; Catheter-directed thrombolysis of deep vein thrombosis enhanced Base on the SIR classification,most of adverse effects were classified as by intraclot microbubbles and ultrasound: A clinical study A to B, no C to F was found.(2) Preoperative and postoperative peak 1 1 2 1 3 Q. Zhu , S. GAO , G. Dong , M. Guo , F. XIE systolic velocity of left and right renal artery were no statistic differen- 1 Department of Ultasound, XinQiao Hospital,Third Military Medical ce(P=0.635,P=0.688).The comparison between baseline and 1,6 months 2 University, Chongqing, China; Department of Ultrasound, The First of blood urea nitrogen, serum creatinine and glomerular filtration rate Affiliated Hospital of Zhengzhou University, Zhengzhou, Henan, China; were no statistic difference (P=0.772, P=0.652, P=0.366). (3) The systolic 3 Internal Medicine Cardiology, University of Nebraska Medical Center, blood pressure dropped 21.5,23.3,22.4mmHg (P=0.000), diastolic blood Omaha, NE, China pressure dropped 11.1,12.9,12mmHg (P=0.000).24-hour systolic blood Correspondence: Q. Zhu pressure dropped 13.6, 15.2, 14.3mmHg (P=0.000),24-hour diastolic Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O49 blood pressure dropped 5.5,6,4.4mmHg(P=0.000); and types of drug dropped 0.8,0.9,1(P=0.000) in the later 1, 3, 6 months. OBJECTIVES This study is aimed to investigate the safety and effective- CONCLUSIONS HIFU can be safely and feasible used in the treatment ness of treating deep vein thrombosis (DVT) using catheter-directed ther- of resistant hypertension. However, further study about long-term apy (CDT) combined with intraclot microbubble-enhanced ultrasound safety and efficacy of HIFU is still needed. therapy (IMUT). METHODS Fourteen patients with acute DVT (<14 days) undergoing CDT were consented to accept coordinated IMUT treatment as the ex- O47 perimental group. During CDT process, percutaneous therapeutic ultra- Efficacy and influential factors of focused ultrasound therapy for sound (TUS) and trans-catheter injection of SonoVue® microbubbles were NNEDV 1,2 simultaneously performed for about 20-30 minutes twice a day depend- C. Li ing on the length of thrombus. A TUS device (SL-10 Sonolyser, Welld The College of Biomedical Engineering, Chongqing Medical University, Medical Electronics Co., Ltd., China) equipped with a single-element, non- Chongqing, Chongqing, China; Haifu Hospital of the First Hospital focused transducer was used for ultrasound thrombolysis. The transducer Affiliated Hospital, Chongqing Medical University, Chongqing, China was operated at the frequency of 1.0 MHz with the duty factor of 0.01 Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O47 and the peak negative pressure was 750kPa or 1.0 MPa depending on the depth of the thrombus. One vial (5mL) of SonoVue® microbubbles OBJECTIVES To investigate the effectiveness and safety of focused into 15 mL saline was infused constantly into the catheter during the ultrasound for treating non-neoplastic epithelial disorders of vulva treatment. Figure 1 shows the working parameters of the therapeutic (NNEDV) and to analyse the factors that affect the effectiveness of fo- ultrasound device. The time frame could be seen in Fig. 2. The other sixty cused ultrasound. acute DVT patients treated with the same CDT procedure without com- METHODS 136 patients with pathologically confirmed NNEDV under- bining IMUT were retrospectively reviewed for the treatment days and went focused ultrasound treatment. Patients were followed up on a overall urokinase dosage as the control group. The criteria for terminating regular basis after treatment. The efficacy was evaluated based on thrombolysis and extubation were vessel recanalization confirmed by degrees of vulvar itching, physical signs and changes in pathological contrast-enhanced ultrasound (CEUS). The major complications like finding in local lesion. The relation between age, course, status of hemorrhage were monitored. The average treatment days and overall menopause, pathological type and curative was analyzed. urokinase doses of the two groups were compared by Independent Sam- RESULTS The average follow-up period was 23.8 months (range 3 ple Test. months to 60 months). The complete remission occurred in 68 of 136 patients. The cure rate was 50% (68/136). 59 were found much RESULTS Images of a patient in the experiment group were showed more improved. The effective rate was 93.38% (127/136). 9 were inef- in Fig. 3. The average treatment days of the experiment group (4.2 fective and the inefficiency was 6.6% (9/136). 7 cases recurred and ± 1.5 d) were significantly less than that of the control (11.8 ± 4.4 the recurrence rate was 5.51%(7/127). No severe side effects were d) (p<0.01). Also, the overall dosage of urokinase used in the ex- found during treatment and no complications were observed during periment group (3.45 ± 1.66 million IU) dropped significantly about follow-up. The age, course of disease and status of menopause were 28.2% when compared to the control group (4.80 ± 2.47 million IU) related to the efficacy (c2=21.017, P= 0.000; c2=26.591, P= 0.000; (p<0.01). Figure 4 presents the results between the two groups. No c2=8.199, P= 0.000). There was no significant difference in the effi- intracranial and local hemorrhage events happened in both groups. cacy of different pathological types (c2=1.635, P= 0.442). CONCLUSIONS By combining IMUT in CDT treatment of acute CONCLUSIONS NNEDV can be treated with focused ultrasound ef- DVT, the treatment days and overall urokinase dosage were re- fectively and safely. The course of the disease, menopause status and markably reduced. This method may help to short hospital stay the age of the patients can be considered as the predictive factors. and reduce the risk of hemorrhage. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 31 of 122 Fig. 4 (abstract O49). See text for description O50 Influence of "T2-RIM Sign" on immediate therapeutic responses to magnetic resonance-guided high-intensity focused ultrasound ablation of uterine fibroids 1 2, 3 3, 4 3 4, 5 5 S. Yeo , Y. Kim , H. Lim , H. Rhim , S. Jung , N. Hwang Radiology, University Hospital of Cologne, Cologne, Germany; 2 3 Radiology, Mint Hospital, Seoul, Korea; Radiology and Center for Imaging Science, Samsung Medical Center, Seoul, Korea; Health Sciences and Technology, SAIHST, Sungkyunkwan University, Seoul, Korea; Biostatistics and Clinical Epidemiology Center, Research Institute for Future Medicine, Samsung Medical Center, Seoul, Korea Correspondence: S. Yeo Fig. 1 (abstract O49). See text for description Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O50 OBJECTIVES Magnetic resonance-guided high-intensity focused ultrasound (MR-HIFU) is a thermal ablation technique, which is gaining acceptance as an alternative treatment option for pa- tients with symptomatic uterine fibroids. To date, previous studies have shown that different MRI characteristic of the uterine fi- broids, such as high overall signal intensity on T2-weighted MR images and high perfusion, can be correlated with poor thera- peutic outcome. Here, we present a study investigating the influ- ence of a high-signal-intensity peripheral rim on T2-weighted MR images (i.e., T2-rim sign) on the immediate therapeutic response of MR-HIFU ablation of uterine fibroids. METHODS This retrospective study was approved by the institutional review board, and patient informed consent was obtained for MR- Fig. 2 (abstract O49). See text for description HIFU ablation. In total, 196 fibroids (diameter 6.2±2.6 cm, range 3.1- 13.6 cm) in 123 women (age 43.4±5.0 years, range 26-55 years) who underwent MR-HIFU ablation from January 2013 to April 2016 were included. The presence of a T2-rim sign and its corresponding percent coverage around the uterine fibroids were assessed on T2- weighted MR images acquired on the day of the treatment. The effects of a T2-rim sign on the immediate therapeutic responses (non-perfused volume [NPV] ratio, ablation efficiency [NPV/treatment cell volume], ablation quality [grade 1-5, poor to excellent]) were in- vestigated with univariable and multivariable analyses using GEE (generalized estimating equation) analysis. In multivariable analysis, T2 signal intensity ratio of fibroids-to- skeletal muscle, relative peak enhancement of fibroids, and subcutaneous fat thickness were also considered. RESULTS The presence of a T2-rim sign significantly lowered the NPV ratio (54.0±28.0% vs. 83.7±17.7%), ablation efficiency (0.6±0.5 vs. 1.3 ±0.6), ablation quality (3.1±1.2 vs. 4.2±0.8), (P<.0001). There were significant negative correlations between the percent coverage of a T2-rim sign and the NPV ratio (ρ=-0.4648, P<0.0001), ablation effi- ciency (ρ=-0.5086, P<0.0001), and ablation quality (ρ=-0.5086, P<0.0001). GEE analysis showed that the presence of a T2-rim sign and its corresponding percent coverage were independently signifi- cant for ablation efficiency and ablation quality (P<0.05). CONCLUSIONS Uterine fibroids with a T2-rim sign showed signifi- Fig. 3 (abstract O49). See text for description cantly poorer immediate therapeutic responses to MR-HIFU ablation. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 32 of 122 O51 continuously updated heat source position. In order to verify the ac- The current clinical applications of MR guided focused ultrasound curacy of the FUS model, temperature distributions were both mod- surgery in China elled and measured. The experimental arrangement (see Fig. 1) H. Wang consisted of a cell containing collagen gel sandwiched between two Radiology, Shanghai General Hospital, Shanghai, China disks of acoustically absorbing PVA gel, in a water tank. This gel Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O51 sandwich was exposed to heating from the moving focus (0.5mm lat- eral extent, 0.5-1.5kW/cm peak intensity) from a single element OBJECTIVES To demonstrate the history and the current state of clin- focused transducer (Sonic Concepts, Bothell, USA, 1.66MHz). The ical applications of MR guided focused ultrasound surgery (MRgFUS) circular trajectory (6-8mm Ø) was traced out for 300s with a period -1 in China. of 1s . The temperature in the cell layer at the centre of the exposed METHODS The history and the current state of clinical applications of circle was measured using a fine wire thermocouple (RS Instruments, MRgFUS from 2011, when it was first introduced into China, was Corby, UK). Simulations of this arrangement were performed, reviewed and surveyed. together with a sensitivity analysis of the thermal conductivity, RESULTS MRgFUS had been approved for the treatment of uterine fi- specific heat capacity, and attenuation coefficient of the materials broids by the China Food and Drug Administration (CFDA) at 2013. used. Water parameters were assumed for speed of sound, and PVA The other registered clinical study of MRgFUS for CFDA is for the pal- gel density. Using these physical and biological model calibrations, liation of pain in bone metastasis, which also had been completed at the viability distribution in cell layers exposed to FUS mediated HT 2015. Moreover, MRgFUS is in ongoing clinical trials for the treatment will be predicted, and compared qualitatively to an experimental cell of adenomyosis of the uterus, uterine incision pregnancy, and oste- viability (MTT) assay. oid osteoma in some institutions. So far, more than 500 MRgFUS RESULTS For biological modelling, the implementation of reproduct- treatments were completed in China. Some institutions are going to ive rather than immediate cell death was essential for the growth re- start clinical trials for the MRgFUS treatment of facet joint syndrome, sponse dynamics of treated cells to be correctly captured. Having painful knee arthritis, breast adenoma, prostate cancer and tremor. calibrated the growth and death rates in the simulation with those CONCLUSIONS Although it was introduced into clinic in China just of HCT116 cells, more growth curves could correctly be predicted several years,MRgFUS,the new noninvasive thermal ablation method, computationally for homogeneous HT, and/or RT treatments (see Fig. already demonstrates its huge potential and prospect. 2). Using the experimental arrangement described above, thermal doses in the therapeutic range (50-250 CEM) were achieved in the cell layer within the heated circular area. The simulation of the time- temperature curves was in good agreement with experimental data once the physical parameters had been adjusted (see Fig. 3 left). O52 However, experimental variations in the thermal and attenuation A predictive simulation framework for combined focused properties, and set-up uncertainties between different gel samples ultrasound hyperthermia and radiation treatment modelling at a had a strong influence on the time-temperature profiles and thermal cellular level dose (see Fig. 3 right). The corresponding measured material proper- 1 2 1 3 1 S. C. Brueningk , G. G. Powathil , P. Ziegenhein , J. Ijaz , I. Rivens , ties, and qualitative comparisons of simulated and experimental cell 4 1 1 M. Chaplain , U. Oelfke , G. ter Haar viability data of treated cell-containing gels will be presented. Joint Department of Physics, The Institute of Cancer Research, Sutton, CONCLUSIONS This CAM presented can easily be adapted to different 2 3 4 UK; Swansea University, Swansea, UK; Bristol University, Bristol, UK; St. cell lines and treatment scenarios. It therefore has potential for use in Andrew's University, St. Andrews, UK future studies of more realistic, tumour-like cell populations (e.g. spher- Correspondence: S. C. Brueningk oids), and their response to FUS and RT. Such simulations may help to Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O52 identify and optimise treatment schedules & exposure conditions. OBJECTIVES Combined radiotherapy (RT) and hyperthermia (HT) treat- Reference ments offer great potential for the successful treatment of radiation- [1] G. Powathil et al., Semin Cancer Biol, (30, p.13–20), 2015 [2] B. Clarke, resistant tumours by thermo-radio-sensitisation. Focused ultrasound (FUS) Ultrasound Med Biol, (21, p. 353–363), 1995 can be used to induce local HT. For treatment planning, it is essential to quantify the biological effects of such combination treatments. For this purpose, we present a multiscale systems oncology simulation framework. The objectives of this study are (1) to design and simulate in vitro FUS ex- periments, (2) to verify FUS simulation using measured temperature distri- butions, (3) to predict the cellular effects of combination treatments. METHODS A 3D cellular automaton model (CAM), based on the com- putational concepts outlined in [1], was implemented in C++, to model cell populations and their treatment response. Each cell undergoes an individual cycle that regulates its treatment response and proliferation. A separate cell survival model was used to calcu- late surviving fractions for the combinations of radiation and thermal doses delivered. From this, a known proportion of cells would undergo immediate, or reproductive, cell death via mitotic catastro- phe. The CAM was compared to results from experiments designed to characterise the response of HCT116 cells in vitro. Clonogenic as- says, cellular growth curves, and flow cytometry analysis were used Fig. 1 (abstract O52). Experimental arrangement used. A gel sandwich to assess overall survival, growth dynamics, and cycle distribution consisting of a thin (~300μm thick), cell containing collagen gel and two after homogeneous heating and/or irradiation. In particular, the influ- 6mm thick slices of PVA cryo-gel is enclosed in a sterile sample holder ence of the cell kill model used in the simulation (instantaneous vs. filled with degassed cell culture medium. Top and bottom of the sample reproductive cell death) was tested. A linear propagation model of holder were sealed with tight Mylar membranes to minimize beam FUS exposure [2] was implemented. To achieve a heated volume sig- reflection. The sample holder is placed with the cell layer located at nificantly larger than the geometric focus, a transducer was moved the focal plane of a focused single element transducer in a water tank. at constant speed in a circular trajectory. The majority of the heated For treatments, the transducer is moved on a circular trajectory with cells are not exposed to FUS directly, but are heated by diffusion of temperature being monitored by a fine wire thermocouple at the thermal energy into the circle’s centre. Heat generation and diffusion center of the circle at the cell layer were simulated by iterative solution of the bioheat equation with a Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 33 of 122 and are different in their density and Young's modulus. The Boundary Element Method is employed to solve the potential flow surrounding the bubble. The bubble is assumed to be adia- batic and initially stable. RESULTS It is seen in Fig. 1 (a) that an oscillating bubble col- lapses with a jet towards the elastic membrane (Sankin & Zhong, 2006). Our simulation result (Fig. 1(b)) compares well to the ex- perimental observationinbubbleshapes and thetimingof events. Figure 2 shows the collapse of the bubble near the elastic membrane after being hit by a shock wave at 114 μs. It is observed in both experiment and simulation that the bubble splits as the membrane moves towards the bubble. We use our Fig. 2 (abstract O52). Comparison of experimental and simulated model to study the effect of initial bubble size in the bubble growth curves for HCT116 cells treated with different modalities. shock waves interaction near fat tissue. It is seen in Fig. 3(a) that Left: 2Gy radiation (22000 cells seeded in 24-well plates, surviving the small bubble of 100 μm in radius collapses with a high speed fraction S2Gy = 0.43(0.38,0.49)). Right: Combination of 2Gy radiation jet towards the fat tissue (and the fat tissue moves towards the and heating for 5min at 46°C (300000 cells seeded in 6-well plates, collapsing bubble). However, if a larger bubble is present (in this S2Gy+5min,46C = 0.11(0.06,0.19)). Cells were simulated with a mean case, 500 μm in radius), the bubble may split before collapsing doubling time of 19.5h. The experimental data points (black circles) (Fig. 3(b)). The fat tissue also moves closer to the bubble just be- are shown, as are the simulated total cell numbers (solid lines), and fore it collapses. We have also examined the effect of shock simulation of the 95% confidence bounds of the surviving fractions wave reflection at rigid interfaces such as bone. Figure 4(a) used (dashed lines) shows the collapse of a 10 μm near bone tissues (on top) after being hit by the lithotripter shock wave. The shock wave is reflected with 0.54 of the original amplitude at the bone inter- face. The bubble collapses at 0.148 μs. Figure 4(b) shows the simulation results without the consideration of the shock wave reflection. The bubble collapses only at 0.166 μs. Thesizeofthe jet is also significantly larger in this case. CONCLUSIONS The interaction between a stationary bubble near various bio-materials and a lithotripter shock wave has been suc- cessfully modeled and simulated using the Boundary Element Method. High speed jets are developed in the bubbles in the direction of travel of the shock waves as previously reported in literature. It is therefore concluded that presence of the bio- Fig. 3 (abstract O52). Comparison of experimental and simulated materials causes the bubble to collapse faster and with slightly time-temperature curves for moving focus (1143 W/cm peak intensity, higher jet speed. However, when the initial bubble size is varied, 6mm diameter of the circular trajectory). Left: The logged temperature different bubble dynamics is observed. In the case of very large data (blue) is first smoothed using a moving average (red), and then bubble (500 μm in radius), the bubble does not collapse with a -1 -1 simulated (yellow) using a thermal conductivity of 0.49Wm K ,aspecific jet towards the bio-material, but will split into smaller bubbles. -1 -1 heat capacity of 2000Jkg K , and an acoustic absorption coefficient of The reflection of the shock wave near rigid boundary such as 0.49nepers/cm at 1.66MHz. Right: Comparison of the simulation (solid bone or stone has the effect of causing the bubble to collapse yellow line) with repeat measurements in the same experimental set-up faster with a narrower jet. for different PVA/gel samples (smoothed data, dashed lines) illustrating experimental uncertainties in thermal dose (15-142CEM) Reference Sankin, G. N. & Zhong, P. (2006), ‘Interaction between shock wave and single inertial bubbles near an elastic boundary’, Phys. Rev. E 74, 046304. O53 Lithotripter shock wave interaction with a bubble near various bio-material 1 1 2 3 S. Ohl , E. Klaseboer , A. Szeri , B. Khoo Institute of High Performance Computing, Singapore, Singapore; 2 3 University of California, Berkeley, Berkeley, California, USA; National University of Singapore, Singapore, Singapore Correspondence: S. Ohl Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O53 OBJECTIVES The interaction of a lithotripter shockwaves with a gas bubble which is located near bio-materials such as fat, skin, muscle, cornea, cartilage, and bone is studied using numerical simulation. Our model is verified with comparison to experimen- tal observations from Sankin & Zhong (2006) where a bubble col- lapses near an elastic membrane after it interacts with a lithotripter shock wave. We proceed to perform a systematic study of the various factors influencing the bubble collapse, in- cluding wave profile, initial bubble size, and reflection of the shock waves at rigid interface (such as bone). The results may be Fig. 1 (abstract O53). Interaction of an oscillating bubble near an useful in the design of lithotripster shock wave therapies and the elastic membrane. (a) Experimental results from Sankin & Zhong prevention of collateral damages in medical treatments. (2006). (b) Numerical simulation of the bubble collapse. Both bubble METHODS The shock waves are modeled as a traveling pressure shapes and timing of events compare well with experiment wave across the domain. The bio-materials have linear elasticity Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 34 of 122 O54 Three-dimensional passive acoustic localization and mapping for cavitation: a preliminary study S. Lu, X. Du, M. Wan Biomedical Engineering, Xi’an Jiaotong University, Xi’an, China Correspondence: S. Lu Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O54 OBJECTIVES Passive acoustic mapping (PAM) based on a linear array has been widely applied in real-time monitoring of high intensity focused ultrasound (HIFU) therapy. By applying a beamforming algorithm to the passively received acoustic emis- sion of cavitation, the obtained spatial distribution of cavitation can be used for prediction of HIFU lesions. However, two- dimensional (2D) PA. M can only provide a plane distribution of cavitation, which is not conducive to clinical diagnosis. PAM based on a hemispherical array Fig. 2 (abstract O53). The interaction of a laser-generated bubble has been used for three-dimensional (3D) vascular imaging with high near an elastic membrane after being hit by a lithotripter shock resolution in the brain, but it is not suitable for treatment monitoring wave at 114 μs. The numerical simulation results are overlapped on of other biological tissues, such as liver and kidney. This means that top of the experimental observations from Sankin & Zhong (2006). It 3D PAM based on an area array for omnibearing monitoring of ultra- is seen that in both simulation and experiment, the bubble splits as sound therapy is required. The objective of this work is to develop a the membrane moves towards the bubble three-dimensional super-resolution passive imaging technique for microvessel and an omnibearing monitoring of ultrasound therapy in real time. METHODS A multi-bubble model was used to create acoustic emissions of cavitation source for single bubble and multiple bubbles. The emissions were recorded by a 32 × 32 area array with an aperture size of 38.4 × 38.4 mm. For three-dimensional (3D) localization of single cavitation bubble, the differential times of arrival between elements at various positions and a reference element was firstly calculated by using cross- correlations. Then a paraboloid function derived by Fresnel ap- proximation was used to fit time-delayed curved surface. Through least square fitting, the estimated paraboloid coeffi- cients were used to calculate 3D position of single cavitation Fig. 3 (abstract O53). A gas bubble collapses near fat tissue (on top) source. For passive mapping of extended cavitation region, a when it is hit by a lithotripter shock wave from below. (a) The initial 3D passive beamforming algorithm based on time exposure bubble size is 100 μm in radius. The bubble collapses with a jet acoustic (TEA) algorithm was applied to the 3D prebeamformed towards the fat interface. The fat tissue moves towards the bubble. (b) data to generate 3D cavitation images. In the algorithm, the The initial bubble radius is 500 μm. The bubble may split before source energy at each imaging location was calculated by inte- collapsing. The fat tissue moves more towards this bigger bubble grating the square of the source strength (delay-and-sum of the prebeamformed data) over a time interval. RESULTS Using a paraboloid to fit time-delayed curved surface can accurately localize single cavitation bubble at different po- sitions in three-dimensional (3D) space, which can be used for super-resolution passive imaging of microvesssel. 3D Cavita- tiom images of single bubble at (0 mm, 0 mm, 40 mm) has a full-width at half-maximum of 0.27mm × 0.27 mm × 2.03mm, which can be regarded as the point spread function at a fixed position of single cavitation source and given parameters of area array (Fig. 1). 3D cavitatiom images and its cross sections along three axes for multiple bubbles (distributed spatially with a normal distribution, the standard devi- ation of the distribution is 0.5 mm × 0.5 mm laterally and 1 mm axially) with different number (N = 20 and 50) demonstrated the feasibility of using 3D TEA-PAM to map extended cavitation region (Fig. 2). The “X-type” ar- Fig. 4 (abstract O53). The effect of the shock waves reflection on tifacts in 3D cavitation images were caused by multiple bub- the collapse of a bubble near bone tissue. The 10 μm bubble is bles interfering with each other, there were no actual bubbles located 10.5 μm from the bone interface. (a) The reflected shock in the artifact region. The artifacts were needed to be im- wave is implemented as a downwards traveling wave 0.54 proved by other technique such as adaptive beamformer. And amplitude of the original lithotripter shock wave after it hits the it should be noted that the lateral resolution of 3D cavitation bone interface. The corresponding time for each bubble shape images is significantly better than axial resolution (Figs. 3, 4). (from outer to inner) is 0, 0.123, 0.137, 0.142, 0.146, and 0.148 μs. The results can assist in real-time 3D monitoring of ultrasound (b) Simulation results without shock wave reflection. The bubble therapy. collapses only at 0.166 μs. For the dotted line bubble shapes, time CONCLUSIONS 3D PAM based on TEA has the potential of providing from outer to inner are 0.154, 0.160, 0.166 μs a novel method for 3D real-time monitoring of ultrasound therapy. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 35 of 122 Fig. 3 (abstract O54). The slices of cavitation images using 3D TEA-PAM algorithms along (a) x axis, (b) y axis, and (c) z axis for multiple bubbles. The cross sections of cavitation images along (d) x axis, (e) y axis, and (f) z axis. The nominal distance from cavitation source to the area array is 40 mm. The number of cavitation bubble is 20, distributed spatially with a normal distribution Fig. 1 (abstract O54). Time delay of the 3D prebeamformed data from single cavitation bubble and the corresponding fitting curved surface. The position of bubble is (a) (0 mm, 0 mm, 40 mm), (b) (0 mm, 0 mm, 80 mm), (c) (9.6 mm, 0 mm, 40 mm), (d) (0 mm, 9.6 mm, 40 mm), (e) (0 mm, -9.6 mm, 40 mm), and (f) (-9.6 mm, 0 mm, 40 mm), respectively Fig. 4 (abstract O54). The slices of cavitation images using 3D TEA-PAM algorithms along (a) x axis, (b) y axis, and (c) z axis for multiple bubbles. The cross sections of cavitation images along (d) x axis, (e) y axis, and (f) z axis. The nominal distance from cavitation source to the area array is 40 mm. The number of cavitation bubble is 50, distributed spatially with a normal distribution O55 Towards a robust and general ultrasound propagation model for MR HIFU Tumour treatments: the hybrid ray tracer approach 1 2 3 4 3 D. Modena , L. Sebeke , M. Baragona , A. Elevelt , R. Maessen , 1 1 D. Bošnački , H. Ten Eikelder Eindhoven University of Technology, Eindhoven, Netherlands; 2 3 University Hospital Cologne, Cologne, Germany; Department of Multiphysics & Optics, Philips Research Eindhoven, Eindhoven, Netherlands; Department of Oncology Solutions, Philips Research Eindhoven, Eindhoven, Netherlands Correspondence: D. Modena Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O55 OBJECTIVES Magnetic Resonance-guided high intensity focus ultra- sound (MR HIFU) has shown to have a clinical relevance for the treat- ment of various types of tumours, such as uterine fibroids and bone Fig. 2 (abstract O54). The slices of cavitation images using 3D metastases. MR imaging is used to plan and evaluate the HIFU treat- TEA-PAM algorithms along (a) x axis, (b) y axis, and (c) z axis for ment, whereas MR thermometry using proton resonance frequency single bubble. The position of bubble is (0 mm, 0 mm, 40 mm) (PRF) is used to monitor the procedure. There are different reasons Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 36 of 122 why the development of ultrasound propagation models for MR- HIFU treatment is of interest. Particularly, they allow the improve- ment of pre-operative therapy planning by predicting the location and the amount of energy deposition during treatment as input for temperature evolution calculation. Furthermore, models can over- come the limitations of the PRF thermometry data, which can be ac- quired only in soft tissues and which present a low spatial resolution compared to the dimension of the treatment region. In addition, these models can enable model- based control during treatment. General accepted methods for HIFU propagation are the Far Field Approximation (FFA) and the Angular Spectrum Plane Wave method (ASPW). However, the first mentioned method is valid only when one homogeneous medium is present, and the second one is not applic- able with no-parallel interfaces. Therefore, in this work we present Fig. 1 (abstract O55). The hybrid ray tracer. In green the rays till the and validate a ray tracer method, a fast and flexible (easily adaptable first interface, and in blue the refracted rays in the tissue- mimicking gel to complex geometries, for both soft tissues and bone cases) ap- proach, which will be used as a baseline for an extension towards a patient-personalized HIFU propagation model. METHODS Firstly, the hybrid stochastic ray tracer is presented (see Fig. 1). The model is suitable for the geometry of the 256-element phased array transducer (focal length 14 cm, frequency 1.2 MHz) used in the MR-HIFU system (Sonalleve V2, Philips, Vantaa, Finland). In the model, ultrasound propagation is represented by rays of inten- sities. The word ‘hybrid’ highlights the fact that the rays leave each transducer element in random directions with an initial intensity de- rived from the FFA formula. The rays follow the laws of refraction and reflection at an interface between two different media. The inter- ference of ultrasound waves in the focal region is calculated from the phase assigned to each ray and the output of the model is the produced power density, which is represented by a 3D table with 0.2 mm grid size. Secondly, our model is compared with the ASPW method in terms of power produced in the focal region, in a config- uration with two propagation media (lossless oil and tissue mimick- ing gel) with only flat surfaces (see Fig. 2). Thirdly, experimental validation data are acquired. In the experimental set-up the ultra- sound waves travel through the lossless material and are focused in Fig. 2 (abstract O55). The ASPW and the hybrid ray tracer outputs the tissue mimicking gel containing Zerdine (CIRS Model 054GS Gen- are compared in a configuration with two homogeneous materials, eral Purpose Ultrasound Phantom). From the PRF thermometry, the separated by a flat interface temperature increases in the coronal slice crossing the focal point are recorded (see Fig. 3). Finally, for this configuration the hybrid ray tracer and the ASPW are used to compute the power density, which then functions as source term in the heat equation. The solution of this equation, found by a finite element approach using COMSOL Multhiphysics (COMSOL, Inc., Burlington, MA), can be compared with the experimental data. However, since the thermal parameters of the tissue- mimicking gel are unknown, these parameters are first found by fitting the results of the ray tracer/ASPW-heat equation to the experimental data. In particular, the diffusion coefficient D=λ /(ρc )(where λ stands for t p t the thermal conductivity, ρ the density, c the specific heat at constant pressure) has been fitted by considering temperature data from the cool- ing down phase, and the term β=b/(ρc ) (where b represents the ab- sorption coefficient, that is the fraction of power produced which contributes to the temperature increase) has been fitted taking into ac- count the heating up phase. RESULTS The hybrid ray tracer predicts a power density in good agreement with the reference method ASPW. After fitting the ther- mal parameters, the resulting model temperature prediction are in accordance with the experimental data. Moreover, the thermal pa- rameters found by the fitting, are in the expected range. CONCLUSIONS The hybrid ray tracer is a good starting point to model MR HIFU treatments, it has been developed to overcome the limitations of the other methods for ultrasound propagation, in particular the prob- lematic treatment of no-flat surfaces. Moreover, through the computa- tion of shear and longitudinal waves in the solid material, the method can be applied to the case of the treatment of bone metastases. Ultim- ately, the hybrid ray tracer discloses advantages such as flexibility and high calculation speed, which make this model a suitable first approach Fig. 3 (abstract O55). The coronal planes and the sagittal plane towards a patient- specific and general HIFU propagation model. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 37 of 122 O56 predict the time-varying ultrasound field generated by HIFU trans- Experimental validation of full-wave HIFU simulations in ducers in heterogeneous, absorbing media under both linear and heterogeneous media nonlinear conditions. These models are likely to find many applica- E. Martin, J. Robertson, B. Treeby tions, from simple in silico investigations, through to patient specific Medical Physics and Biomedical Engineering, University College London, treatment planning. It is important to note, however, the close agree- London, UK ment relies on accurate knowledge of the geometry, position, and Correspondence: E. Martin material properties of the heterogeneities in the beam path. While Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O56 this is possible in a laboratory situation, the sensitivity of the simula- tion output to these parameters is not yet fully understood. Further OBJECTIVES Numerical simulations have a wide variety of applica- work is needed to clarify the uncertainties in numerical predictions tions in high-intensity focused ultrasound (HIFU), including trans- using real biological tissue where the geometry and material proper- ducer design, patient selection, treatment verification, and treatment ties may not be precisely known. planning. However, for these simulations to be clinically useful, it is crucial that the accuracy of the numerical model is rigorously estab- lished under realistic conditions. The aim of this study was to quanti- tatively validate the wave models within the open-source k- Wave O57 Toolbox using a series of experimental measurements made with Translating microbubbles with millisecond scale ultrasound pulses: heterogeneous fluid and solid objects in the beam path. implications for controlled transport of bubbles to a boundary 1,2 2 1,2 METHODS Experiments were performed in a 60 x 60 x 100 cm tank C. Acconcia , A. Wright , D. Goertz 1 2 of temperature controlled, degassed, and deionised water. The University of Toronto, Toronto, Ontario, Canada; Sunnybrook Research acoustic field was generated by a single-element spherically focused Institute, Toronto, Ontario, Canada Correspondence: C. Acconcia HIFU transducer driven at 1.1 MHz with a 4 cycle burst. Signals were acquired using a calibrated 75 μm needle hydrophone connected to Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O57 a 5- axis computer controlled positioning system. Planar scans were acquired both parallel and perpendicular to the beam axis following OBJECTIVES To elicit bio-effects, ultrasound (US) stimulated micro- bubbles (MBs) must be in close proximity to their target. In a situ- propagation through the heterogeneity. Measurements were made at several drive levels to give both linear and nonlinear conditions. ation such as sonothrombolysis, however, the majority of circulating Two types of medium inclusions were used to obstruct the beam MBs will not be in proximity to the clot boundary. Radiation forces can path. The first was a series of fluid filled containers (rectangular, potentially direct MBs to clot surfaces and recent work (Acconcia et al., wedge, and planoconvex) constructed from laser cut Perspex and a JASA 2016) has suggested clot degradation patterns (Fig. 1A) could be stretched Mylar membrane. These were filled with either olive oil or influenced by ‘transport pulses’. However, this is a complex process in- glycerol, and were intended for validation of the models within a volving a population of MBs flowing in a vessel, with size-dependent ra- soft-tissue like medium. The second was a series of 3D printed and diation and drag forces. We recently conducted experiments to resin cast skull bone phantoms derived from a T1 MR image. These investigate the process of bubble accumulation at a boundary with a were intended for validation of the models when heterogeneities view to improving exposure strategies in applications such as sono- with a larger acoustic impedance mismatch are present. The trans- thrombolysis. The size and spatial distribution of bubbles arriving at a ducer and medium inclusions were rigidly positioned using a series fibrin clot boundary (Fig. 1B) were found to be highly dependent on of custom-made Perspex and 3D printed mounts and opto- pressure and flow conditions. The ability to model and control this mechanical components attached to a breadboard mounted over process is limited by a paucity of data on the translational dynamics of the water tank. This allowed accurate registration of the source and encapsulated MBs under the influence of millisecond scale US pulses. object positions. For each experiment, a numerical simulation was To date, experimental work for individual MBs has focused on the use conducted using the MPI version of the open-source k-Wave Toolbox of shorter (imaging) pulses. The purpose of this work is to directly in- running on the IT4I Salomon supercomputer. The grid parameters vestigate the translation of individual MBs with millisecond pulses in used 6 spatial points per minimum wavelength for the fluid objects, order to constrain theoretical modeling of this process. and 10 points for the solid objects. The source conditions were METHODS In house developed optical tweezers integrated with a established using free- field measurements and linear acoustic holog- microscope (60x) and high speed camera were employed to select indi- raphy. For measurements at other drive levels, the source pressure vidual Definity MBs and position them in a fluid region away from was assumed to scale linearly with drive voltage. The medium prop- boundaries. MBs were then subjected to a series of 1 ms length, 1 MHz erties within the simulation were specified according to the known pulses and were recorded (10 kframes/s) while translating laterally position and geometry of the scattering object, and using book through the optical field of view. Image analysis was employed to values for the material properties. The measured and simulated wave quantify the displacements as a function of time with the primary fields were compared using several metrics, including the peak posi- metric of interest being the distance travelled within the first pulse. Ex- tive and negative focal pressure, focal volume, focal position, arrival posures were conducted at transmit pressures of 25, 50, 100, 150 and time of reflections, and L2 error in the spatially varying wave field. 200 kPa. For each pressure a range of bubble sizes (1.5-9 microns in For nonlinear conditions, the fields after spectral decomposition were diameter) were assessed (n=86). The radial oscillations of encapsulated also compared. MBs were calculated based on a variant of the modified form of the RESULTS The simulations and experiments showed close quantitative Rayleigh-Plesset equation proposed by Marmottant et al (JASA 2005). agreement. Errors were typically < 3% for the peak positive pressure, This model requires as inputs estimates of shell elasticity, dilitational < 3% for the focal volume, < 1 mm for the focal position, < T/12 for viscosity, and a starting effective surface tension. To assess displace- the arrival times, and < 6% for the L2 error. The exception was the ment, radiation force along with other relevant forces on the MB measurements made using the resin cast bone phantom, where sim- (added mass, quasi-steady drag and history force) were solved numer- ulations overestimated the focal volume by ~12%. This discrepancy ically. The history force integral was evaluated using methods pre- is likely due to variations in the material properties of the resin from sented by Garbin et al. (2009), and Chung et al. (1982). After empirical the casting process, which were not captured by the numerical modification, the history force integral has been generalized to apply model. For the nonlinear conditions, errors were smallest at the over a larger range of Reynolds numbers. For this finite Reynolds num- fundamental frequency, but still remained acceptable at the fifth ber form of the history force (Takemura et al, JFM 2004), we used the harmonic (the highest harmonic measured with significant signal- numerical integration scheme proposed by Chung et al. (1982). to-noise). RESULTS The displacement as a function of MB size and pressure are CONCLUSIONS The results demonstrate that the full-wave models in shown in Fig. 1C. A pronounced feature of the displacement curves the open-source k-Wave toolbox can accurately and quantitatively is the presence of an effective threshold point in size, below which Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 38 of 122 there is only minimal translation. The threshold size increases with OBJECTIVES To reduce hyperthermia treatment times, there has been decreasing pressure. Note that single pulse displacements higher a tendency to operate HIFU in short bursts, at high power levels. This than 140 microns were not observed due to field of view constraints. necessitates the introduction of laboratory thermometry methods that The threshold dependant nature of the onset of rapid displacement are temporally and spatially resolved, and are able to characterize the could be well accounted for by the threshold dependant nature of burst of thermal output in real time. In this work we describe the devel- oscillations captured by the encapsulated MB model. Example mod- opment of a novel optical imaging method for temperature measure- eling is shown for the 25 kPa case (Fig. 1D), with a local maximum ments in a HIFU field that is fast and spatially resolved. The method is about a peak size, which correlates with the approximate resonant non invasive and only requires optical access to operate. size of Definity at 1 MHz. The model results for up to 100 kPa were METHODS The described thermometry method is based on the found to be in good agreement with data, when employing shell pa- temperature-dependence of the optical refractive index in solids, rameters that were within the range of those previously reported for gels, and liquids. As an example, water and polyachrylamide gel have -4 phospholipid agents. Importantly, the inclusion of history force was a linear dependence of around k=-1x10 /°C. When a laser ray travels required to accurately fit the data. through an axisymmetric HIFU heated spot, it will deflect from its CONCLUSIONS This study reports the first size dependant data set straight unperturbed path. The net angular ray deflection is for the translation of MBs under the influence of ms scale US pulses measured and converted to temperature using a direct formula at therapeutically relevant pressures. A key feature was threshold size derived from solving the paraxial eikonal ray equation for a Gaussian dependent displacements, where the degree of translation was phase object. In the experimental setup, a 65 mm aperture HIFU highly sensitive to the radial asymmetry of the oscillations inherent transducer (Sonic Concepts) running at its third harmonic of 1.6 MHz with this model, a behavior that is characterized by compression was focused on an optically transparent tissue mimicking phantom dominated oscillations for smaller bubbles (<~3-4 microns). The ma- cube of 2 cm side length placed in a water bath for acoustic jority of simulations to date have excluded history forces on the basis coupling. A laser light sheet of 3 cm height was passed through a of Reynolds number arguments, however for the conditions investi- custom made comb to chop it into individual rays producing a gated here the inclusion of history force was found to be important. planar bundle formation. The ray bundle was then focused with a When this constrained MB model was applied to our previously ac- cylindrical lens to allow as many rays as possible to pass through the quired data set of size dependant MB arrivals at a planar surface, 0.3 mm heated spot to improve the spatial resolution. The acoustic good agreement was found, thereby providing a tool to investigate and optical axes were orthogonal. The ray bundle was imaged at 200 and optimize the control of translating MBs to a surface. fps, at a location around 25 cm down the laser path from the HIFU spot (providing geometric advantage for imaging) during HIFU irradiation to capture the heating and cooling phases. The ray deflection angles from the individual images (relative to no HIFU image) were extracted and converted into radial temperature profiles. The tissue mimicking phantom cube was made by casting liquid silicone that cured in 24 hrs. K- type thermocouples (TC) of 0.13 mm dia. were cast within the phantom to read temperature. Figure 1 shows a picture of the HIFU installed in the experimental setup and a sample raw image of the ray bundle at the imaging location. Geometric triangulation was used to map the rays from the imaging plane to the HIFU focal plane. RESULTS By comparing sequential ray images, ray deflection maps can be extracted as shown in Fig. 2. The resulting radial distribution of the angular deflection profile at the HIFU focal plane is also shown in the figure. The radial deflection profile is directly substituted in the aforementioed equation (shown in Fig. 2) to produce radial temperature profiels at different time steps as shown in Fig. 3. In this figure, the HIFU was operated to produce a 2 ms burst at a power level of around 20 W. As noted, the optical imaging thermometry is able to resolve the temperature distribution across the 2 mm HIFU heated spot giving a spatial resolution better than 0.1 mm. The temperature profile is narrow with a high peak right after HIFU ex- posure. The profile widens and cools off with time due to heat diffu- sion. In Fig. 4 the temporal evolution of the peak in the temperature distribution is plotted alongside the TC reading. The TC signal spikes Fig. 1 (abstract O57). A) Two-photon microscopy of the erosion higher than the optical ray temperature signal, expectedly due to vis- zone of the flourescently tagged fibrin network of a treated blood cous heating. The two curves agree well during the latter cooling clot. Upon arrival, MBs penetrate and disrupt the fibrin network. B) Top phase. The temporal resolution of the method is fixed by the frame view of the size dependant arrival of MBs at a planar fibrin clot boundary. rate of the camera which in this case is 0.02 ms. Faster frame rates A minimum intensity projection is shown over 100 US pulses. C) Size are straightforward to obtain, but in this case were limited by the and pressure dependant displacement of MBs from a single 1 ms available laser power illumination. pulse. D) A comparison of data and modeling for the 25 kpa case CONCLUSIONS A fast rise-time and spatially resolved optical imaging ray-bundle thermometry method has been developed and demon- strated with milliseconds long HIFU bursts in a tissue mimicking phantom. Unlike previously proposed thermometry methods such as O58 laser-induced fluorescence which mainly works in liquids, the current A fast non-invasive optical imaging thermometry method for HIFU ray bundle method works equally well in liquids and solid gels and it G. Oweis, H. Daoud does not require the tedious calibration steps. It requires working Mechanical Eng, American University of Beirut, Beirut, Lebanon with optically transparent materials. The simplicity of the experimen- Correspondence: G. Oweis tal setup and ease of image processing, in combination with the Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O58 Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 39 of 122 availability of lasers and cameras in most laboratories makes this method a viable choice for fast and resolved characterization of HIFU thermal outputs. Fig. 1 (abstract O58). Experimental setup showing the HIFU in the water bath, the focused laser bundle illuminating the tissue phantom, and imager (left); and a sample raw ray bundle image in the imaging plane (right) Fig. 4 (abstract O58). Temporal evolution of the peak temperature at the center of the heated spot using the optical ray method (black) alongside thermocouple readings (green) O59 In vivo and ex vivo monitoring of thermal ablation in a porcine model using ultrasonic nakagami imaging S. Zhang, S. Shang, Y. Han, R. Xu, C. Gu, L. Zhang, M. Wan Department of Biomedical Engineering, Xi'an Jiaotong University, Xi'an, Shaanxi, China Correspondence: S. Zhang Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O59 Fig. 2 (abstract O58). Two dimensional ray deflection map in the imaging plane (left); and the corresponding radial distribution (r) of OBJECTIVES The uses of high-intensity focused ultrasound (HIFU) the angular ray deflections in the HIFU focal plane (right); also and microwave ablation (MWA) as a non- invasive or minimally inva- shown is the temperature conversion equation sive therapeutic technique are being investigated due to the devel- opment of the monitoring imaging techniques. The acoustic posterior shadowing effects of cavitation and/or boiling bubbles influence the accuracy for defining the location of thermal lesions when using ultrasonic monitoring imaging. This in vivo and ex vivo study investigated the feasibility of using ultrasonic Nakagami imaging to evaluate the thermal lesions during HIFU and MWA in a porcine model. METHODS 2-D RF data backscattered from the ablated region were captured by a modified diagnostic ultrasound scanner to estimate ultrasonic Nakagami parameters of the thermal lesions, and to recon- struct the ultrasonic B-mode and Nakagami images during HIFU and MWA. A term contrast-to-noise ratio (CNR) between the thermal le- sions and the surrounding normal tissue is used to estimate the con- trast resolution of the ultrasonic B-mode and ultrasonic parameter images. RESULTS Unlike Ultrasonic B-mode images, Nakagami images were less affected by the shadow effect in monitoring of thermal ablation, and a fairly complete hyper-echoic region was observed in the Naka- gami image. After thermal ablation, a bright hyper-echoic region ap- peared in ultrasonic Nakagami parameter images as an indicator of the thermal lesion. Mean values of the Nakagami parameter in the thermal lesion region increased to 0.58, 0.71 and 0.91 after 1, 3 and 5 min of thermal ablation. CNR values calculated for Nakagami par- ameter images increased from 0.13 to approximately 0.61 during thermal ablation and then decreased to 0.26 at the end of the post- Fig. 3 (abstract O58). Radial temperature distribution of the HIFU ablation stage. The corresponding CNR values calculated for the heated spot at increasing times from the HIFU pulse ultrasonic B-mode images were 0.24, 0.42 and 0.17. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 40 of 122 CONCLUSIONS This in vivo and ex vivo study on a porcine model suggested that the Nakagami parameter may have the potential use to evaluate the formation of thermal lesions and the ultrasonic Naka- gami imaging may provide an alternative modality for monitoring HIFU and MWA treatment. O60 Changes in the optical scattering and absorption spectra of ex-vivo chicken breast tissue following exposure to HIFU J.L. Raymond, R. Cleveland, R.A. Roy Department of Engineering Science, University of Oxford, Oxford, UK Correspondence: J.L. Raymond Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O60 Fig. 1 (abstract O60). Optical reduced scattering coefficient (μ ') OBJECTIVES Real-time acousto-optic (AO) sensing has been shown versus optical wavelength (λ) for 10 s HIFU exposures at 150, 200 to non-invasively detect changes in ex vivo tissue optical properties and 250 W/cm during high-intensity focused ultrasound (HIFU) exposures. Baseline changes in optical properties have been previously measured as a function of thermal dose for chicken breast exposed to a temperature- controlled water bath (doi:10.1088/0031-9155/59/13/ 3249). In this work, the wavelength-dependent optical scattering and absorption coefficients of ex vivo chicken breast tissue exposed to HIFU were measured using an integrating sphere spectrophotometric technique. METHODS Thin tissue sections (approximately 2 mm) were mounted on an acoustically transparent membrane such that the bottom sur- face was coupled to a 37°C water bath and the top surface exposed to air to permit non-contact thermal measurements using an infrared camera. Thermal damage was induced using a focused 1.1-MHz transducer (H-102; Sonic Concepts, Bothell, WA, USA) coupled to an acoustic lens and positioned in the water bath below the tissue sam- ple. Phase-shifts produced by the lens de-focused the beam and re- sulted in an annular focal zone with the acoustic intensity maximum located 1 mm off-axis. Thus, a larger focal heating area could be pro- duced in the tissue sample with less spatial variation in temperature (and thermal dose) in the region-of-interest (ROI) than for a tightly focused beam. Spatiotemporal surface temperature elevations were measured using an infrared camera (FLIR Systems, Kent, UK) and used to calculate the spatially-dependent thermal dose delivered to the tissue ROI. Optical property changes in the ROI were measured using a dual-beam UV-Vis-NIR spectrometer (Lambda 750s; PerkinEl- mer, Beaconsfield, UK) equipped with a 100 mm integrating sphere. The exposure intensity and time were varied in order to determine the optical property changes in tissue as a function of the delivered thermal dose. RESULTS Figure 1 plots the optical reduced scattering coefficient Fig. 2 (abstract O60). Optical reduced scattering coefficient (μ ') at (μ ') versus optical wavelength (λ) for 10 s exposures at 150, 200 and 975 nm as a function of the measured thermal dose for all sonications 250 W/cm . In Fig. 2, the reduced scattering coefficient (μ ') at 975 nm is plotted as a function of the measured thermal dose for all sonications. Results show that HIFU-induced thermal damage results in changes in scattering at all optical wavelengths from 400-1300 nm (Fig. 1). Furthermore, the reduced optical scattering coefficient in- O61 creases dramatically for exposures exceeding approximately 10^3 cu- In vivo comparison of ultrasound and magnetic resonance mulative equivalent minutes at 43°C (CEM ) (Fig. 2). thermometry for guidance of HIFU mild hyperthermia 2, 1 2 1 2 1 CONCLUSIONS The apparent threshold for optical property changes R. Staruch , S. Sethuraman , B. Cheng , J. Kruecker , R. Chopra in chicken breast tissue is broadly consistent with other studies of Department of Radiology, UT Southwestern Medical Center, Dallas, the thermal dose threshold for lesion formation. AO monitoring of Texas, USA; Ultrasound Imaging & Interventions, Philips Research North HIFU therapy is feasible and this modality may be useful as an alter- America, Cambridge, Massachusetts, USA native to thermometry and dosimetery. Wavelength-dependent op- Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O61 tical property changes can be used to improve the AO sensing of lesion formation during HIFU therapy. [Work supported by the F. V. OBJECTIVES To determine the in vivo feasibility of using strain-based Hunt Postdoctoral Fellowship of the Acoustical Society of America, ultrasound (US) thermometry to monitor mild HIFU heating in muscle the University of Oxford, and EPSRC grant number EP/K02020X/1]. tissue, by direct comparison of temperature measurements made Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 41 of 122 simultaneously using US strain estimation, magnetic resonance (MR) thermometry, and implanted optical sensors. METHODS US thermometry in thigh muscle of anesthetized rab- bits was performed in a 3T MRI (Ingenia, Philips) using a single- element 5 MHz US transducer (Y109, Sonic Concepts) fitted within the central aperture of a single- element 1.1 MHz HIFU transducer (H102, Sonic Concepts); both had a focal distance of 59 mm. A fiber-optic temperature sensor (T1C, Neoptix) was im- planted into the thigh muscle; its location was visualized using 3D T1- weighted MRI (Fig. 1). Transducer location was controlled using an MR-compatible positioning system (RK100, FUS Instru- ments), to set the US focus at offsets of 0, 2, and 4 mm away from the sensor. US pulse-echo acquisition and HIFU energy de- position were interleaved using an open-architecture US system (Vantage 128 HIFU configuration, Verasonics). Filtered US signals were passed into the MR scan room through a grounded RF Fig. 2 (abstract O61). Agreement between MR thermometry and penetration panel. US echo shifts were tracked in the raw RF US fiber optic sensor in image slices along and across the HIFU focus data to derive thermally-induced strains which are proportional to temperature rise. MR temperature maps were calculated using the proton resonance frequency shift technique from RF-spoiled fast field-echo phase images (echo time 12 ms, in-plane reso- lution 2 mm, slice thickness 4 mm) acquired across and along the HIFU focus. The temporal resolutions of the MR, US, and fiber-optic acquisitions were 5, 1, and 1 seconds, respectively. RESULTS Simultaneous US and MR temperature mapping data was successfully acquired and compared with invasive fiber-optic measurements during 15 HIFU sonications in 2 rabbits. MR temperature measurements in a 4 x 4 mm ROI at the location of the fiber optic sensor agreed well with optical measurements, with mean difference and temporal variation less than 0.5°C (Fig. 2). US thermal strain profiles acquired at the location of the HIFU focus 2 to 4 mm away from the sensor correlated well with sensor readings (R = 0.93 ± 0.03, Fig. 3). CONCLUSIONS In in vivo rabbit muscle under normal respiration and perfusion, strain-based ultrasound thermometry is feasible in the mild hyperthermia range. Fig. 3 (abstract O61). Agreement between scaled US thermal strain and fiber optic sensor in rabbit thigh O62 Changes in backscatter of liver tissue due to thermal heating can be used for guiding focused ultrasound ablations 1,2 1,2 V. Barrere , D. Melodelima 1 2 LabTAU, INSERM, Lyon, France; Université Claude Bernard, Lyon 1, Lyon, France Correspondence: V. Barrere Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O62 OBJECTIVES High-intensity focused ultrasound (HIFU) is a noninva- sive therapeutic modality concentrating ultrasonic energy in a small volume of biological tissues. Only in the focal volume the temperature rises above the threshold of thermal coagulation. A non-invasive modality is needed to effectively guide and monitor non-invasive HIFU treatments. Today, magnetic resonance imaging (MRI) and ultrasonic imaging are the two main modalities used in combination with HIFU for guiding the treatment. MRI is superior to ultrasound in visualizing tissue temperature and necrosis but this technique is highly expensive and lacks portability. Ultrasonic im- aging has advantages in its inexpensiveness and portability. In Fig. 1 (abstract O61). 3D T1-weighted MRI of in vivo experiment addition, in most cases the ultrasound imaging probe is placed such setup for simultaneous ultrasound and MR thermometry in rabbits that the imaging plane is aligned with the HIFU acoustic axis. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 42 of 122 Conventional B-mode imaging shows the spatial distribution in amp- METHODS Collection and handling of postsurgical breast specimens litude of the echoes reflected by acoustic impedance mismatches were approved by the institutional review board (IRB) of Columbia and is widely used to guide and monitor HIFU treatments. However, University. Six post-surgical mastectomy breast specimens were ob- B-mode imaging provides limited information about the formation of tained immediately after surgery for HMI imaging. The HMI setup coagulation necrosis due to HIFU. In most cases hyperechoes are vis- consists of a 93-element, 4.5-MHz HIFU transducer confocally aligned ible due to microbubbles generated by either acoustic cavitation or with a 64-element 2.5-MHz phased array to transmit and receive boiling. One of the limitations is that tissue thermal coagulation is through a 4-board VDAS system. The HIFU transducer was driven by not always linked with microbubble generation. In addition, it is diffi- an amplitude-modulated sinusoidal signal to vibrate the tissue at cult to contour precisely the ablated zone based on these hypere- focal area. To generate a 2D/3D HMI displacement map, a point-by- choes. Several methods have been proposed to characterize thermal point raster scan acquisition was used with a step size of 2 mm. At change based on other parameters, such as ultrasonic backscatter. each spot, the focused ultrasound exposure was 0.06 s long (3-cycle One of the oldest methods is the temperature estimation from the oscillations at 50 Hz), during which 60 RF frames at 1-kHz pulse repe- echo shift due to thermally induced change in speed of sound but is tition frequency were acquired for cross-correlation. limited to temperature up to 50-55°C. Thermally induced change in RESULTS 60x15 mm HMI displacement maps could be generated to ultrasonic attenuation has also been studied and used to monitor HIFU map the relative stiffness on the target area within 15 minutes (Fig. treatment, but the change in the attenuation coefficient is due to co- 1) indicating lower displacement in the tumor region and higher dis- agulation and not to the temperature rise. Techniques for estimating placement in the peripheral tissue. In Fig. 1, the average peak-to- the elasticity of tissue are under also investigation based on the fact peak displacement in the tumor defined on the B-mode was found that tissue become stiffer when coagulated but not as a function of the to equal 6.52±3.65 μm, and 38.70±21.79 μm in the surrounding tis- temperature. In this study we investigated the change in ultrasonic sue. A Student’s t-test showed significant difference (P < 0.0001) be- backscattered energy due to the thermal coagulation itself without tween the tumor and peripheral tissue in the HMI displacement. In microbubble generation from 37°C and up to 70°C. the meanwhile, HMI displacement map showed larger relatively stif- METHODS To minimize the cavitation nuclei in the tissue sample, the fer region compared to tumor region on the B-mode. tissue was carefully degassed prior to HIFU exposure. A total of 26 CONCLUSIONS HMI can successfully map the relative stiffness at vari- experiments were performed in porcine liver. Liver samples were able depths on post-surgical human breast mastectomy specimens used because the knowledge of ultrasonic backscatter is largely with or without the skin. The tumor on the HMI appeared slightly larger described. The origin of ultrasonic backscatter from liver tissue is than the one delineated on the B-mode. This study laid the foundation attributed to collagenous septae between liver lobules. Ultrasonic for future clinical study on HMI guided focused ultrasound treatment. backscatter change due to cell death is attributed to the destruction of the cell nuclei. Ultrasonic RF signals at a center frequency of 2.5 MHz backscattered from the tissue before, during and after thermal coagulation due to HIFU exposure at 3MHz were obtained with a pulse-echo transducer and analyzed off-line. The tissue before, during and after the thermal coagulation was also examined by histology using an optical microscope. These results were then com- pared and discussed to clarify the mechanism of the backscattered energy change due to thermal coagulation induced by HIFU. Long exposure time (120 seconds) was used to observe smooth temperature increase from 37 to 70°C. RESULTS The model predicted a linear increase 10 dB. A linear increase 8 dB was measured in ultrasound backscattered power during experiments. The tissue temperature increase estimated using backscattered energy correlated well (r=0.79) with temperature measurements performed using thermocouples. This linear relation- ship between changes in the backscattered energy and actual temperature was observed up to 70°C. CONCLUSIONS Successful temperature estimation may allow creat- ing 2D temperature maps during HIFU treatments. O63 Tumour characterization of human breast mastectomy specimens using Harmonic Motion Imaging (HMI) 1 1 1 1,2 Y. Han , S. Wang , T. Payen , E. Konofagou 1 2 BME, Columbia University, New York, New York, USA; Radiology, Fig. 1 (abstract O63). (a) B-mode image of a post-surgical breast Columbia University, New York, New York, USA mastectomy specimen with tumor appear dark in the yellow dash Correspondence: Y. Han line. (b) HMI displacement overlay on B-mode image with color bar Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O63 showing the HMI displacement. *The HMI scan was not completed due to time constraints according to the IRB protocol OBJECTIVES Breast cancer is the most common cancer as well as the second leading cause of cancer death among women. There is a need to develop a breast imaging technique for reliable identifica- tion and differentiation of breast masses based on stiffness. Recently we have shown that Harmonic Motion Imaging (HMI) can be used to O64 differentiate relative stiffness and monitor HIFU ablations in small Monitoring of nonlinear scattering during cavitation-enhanced lumpectomy human breast specimens. The objective of this study is ultrasonic heating for coagulation detection to apply HMI on post-surgical mastectomy breast specimen with or S. Yoshizawa, K. Tomiyasu, R. Iwasaki, R. Takagi, S. Umemura without skin to mimic the in vivo environment and characterize Tohoku University, Sendai, Japan tumor at different depth for better tumor localization and identifica- Correspondence: S. Yoshizawa tion before and after HIFU treatment. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O64 Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 43 of 122 OBJECTIVES A noninvasive technique to monitor thermal lesion formation is necessary to ensure the accuracy and safety of high- intensity focused ultrasound (HIFU) treatment. Methods for the co- agulation detection and temperature monitoring using ultrasound echo changes, such as the decorrelation of echo signals, have been investigated. On the other hand, methods for the enhancement of HIFU heating using cavitation bubbles have been investigated for the efficient thermal treatment. However, it is difficult to apply the monitoring methods using ultrasound echo changes during cavitation-enhanced HIFU heating because cavitation bubbles tend to produce random echo signals. The objective of this study is to de- Fig. 1 (abstract O64). Schematic of experimental setup velop a noninvasive technique to detect the thermal lesion formation in cavitation- enhanced ultrasonic heating. In this study, nonlinear components of scattered ultrasound from cavitation bubbles are ana- lyzed and used for the coagulation detection. METHODS Figure 1 shows a schematic of the experimental setup. A degassed chicken breast tissue was used as a target tissue. The water was kept at approximately 36°C. HIFU was generated by a 256- element array transducer (Imasonic) with both diameter and focal length of 120 mm. The transducer was connected to 128-ch staircase voltage amplifiers (Microsonic) by electrically combining each two adjacent elements and driven at 1 MHz. A phased array probe (Hitachi Aloka UST-52105) was set in the central hole of the HIFU transducer and connected to a programmable ultrasound imaging system (Verasonics Vantage 256). Figure 2 shows the HIFU sequence consisting of high-intensity short pulses to generate bubble clouds, named “trigger pulses”,and following moderate-intensity long bursts for the enhancement of the ultrasonic heating, named “heating bursts”. The focal point of the trigger pulse was electronically scanned at each corner of a regular hexagon 3 mm each side and a ring focal region was generated employing a sector vortex method in the heating burst exposure to cover the six foci of the trigger pulse for the volumetric cavitation-enhanced heating. The total acoustic powerforthetriggerpulse andheating burst were 1800 and 90 W, re- spectively. The duration and interval time for trigger pulses at each focal point were 25 and 3 μs, respectively. The trigger pulses were laterally Fig. 2 (abstract O64). HIFU sequence consisting of high-intensity short scanned for four times. For heating bursts, the duration and interval time pulses to generate bubble clouds and following moderate- intensity long for trigger pulses at each focal point were 5 ms and 4 μs, respectively. bursts for the enhancement of the ultrasonic heating The focal spot was scanned 5 times. The subtotal durations of trigger pulses and heating bursts were 0.67 and 50 ms, respectively. Immedi- ately after the end of the heating bursts, a 2-ms interval time was re- served for ultrasonic imaging with planewavetransmissionsata frequency of 1.88 MHz. Ultrasonic RFdatawerealsoacquiredduringthe HIFU exposure for the passive coagulation detection. RESULTS Figure 3 shows temporal change in spectral intensity calcu- lated from the RF data during the HIFU exposure after the receive beam- forming with the fixed focus at the HIFU geometric focus. The blue and red lines denotes the acoustic signal intensity at 1 and 2 MHz, respect- ively. Figure 4 shows pulse inversion (PI) images during the HIFU interval time just after HIFU duration started and at the moment when high brightness appeared, 8.0 s after the start of HIFU exposure. The result showed a good correlation between the intensity of the second har- monic HIFU echoes and the emerging high brightness in the PI image. The emerging high brightness was considered as boiling bubbles. It is inferred that the intensity of the second harmonic HIFU echoes in- creased due to newly generated cavitation bubbles which were caused by the trigger pulse reflected by the boiling bubbles. In other words, when the spectral peak was seen, the treatment region was heated suffi- ciently and the cavitation threshold in the surrounding region was de- ceased because of the relatively high temperature. By stopping the HIFU exposure at this moment, overheating would be avoided. CONCLUSIONS In this study, the nonlinear scattering during cavitation- enhanced ultrasonic heating was monitored with an ultrasound im- aging probe. The intensity of the second harmonic HIFU echo increased at the moment when high brightness appeared in the PI image, indi- cating the coagulation in the HIFU focal region. The results will be Fig. 3 (abstract O64). Spectral intensity calculated from the RF data shown and discussed in the presentation when the HIFU exposure is during the HIFU exposure after the receive beamforming with the automatically stopped by detecting the second harmonic intensity fixed focus at the HIFU geometric focus exceeded the certain threshold. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 44 of 122 recorded. The experiment was repeated with various attenuating layers added to the beam path between the target ROI and the FUS transducer. Four different layers with attenuation between 1.2 to 6.3 dB were used. RESULTS The destruction of microbubbles by the treatment beam increases with treatment amplitude and burst length, as expected (Fig. 2). However, any single measurement cannot reliably reveal the in situ beam intensity without controlling the microbubble concentration. The data sets of destruction curves obtained with attenuating layers were matched to the un-attenuated reference data set using a multi- parameter fitting algorithm (Fig. 3). The resulting fitted beam intensity was found to match closely the actual values, verified by independent measurement of attenuation. The errors of in situ beam intensity mea- sured with the proposed method were found to be less than 1.1 dB. CONCLUSIONS Analyzing the complete destruction characteristics produced a feature-rich data set that can be fitted to more than one unknown parameters simultaneously. High frame rate imaging pro- vides crucial speed advantage to this procedure. We proposed a scheme in which, at a pre-treatment phase, a patient is injected with Fig. 4 (abstract O64). PI images during the HIFU interval time just microbubble contrast agent, exposed to low-dose ultrasound from after HIFU duration started and at the moment when high the treatment device, have the destruction characteristics of the bub- brightness appeared bles analyzed to ascertain ultrasound focus and in-situ intensity, and compensation to the therapy planning applied, before the actual course of treatment is applied. This study also demonstrated some O65 capabilities of the in-house designed 2D array therapy system. Par- Estimation and Compensation of in-situ ultrasound intensity using ticularly interesting is that an arbitrary treatment ROI can be exposed a 2D array therapy system and high frame rate imaging in less than 32 ms. In this study the scanning speed of the treatment 1,2 1,2 1,2 1,2 1,2 1,2 Y. Peng ,B.He , N. Deng , X. Chen , S. Chen , C. Chin focus was exploited to ensure that the entire ROI is exposed evenly Biomedical Engineering, Shenzhen University, Shenzhen, Guangdong, in between excessive imaging events. China; National-Regional Key Technology Engineering Laboratory for Medical Ultrasound, Shenzhen, China Correspondence: Y. Peng Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O65 OBJECTIVES We investigated the potentials for ultrasound to provide some patient-specific information which would improve delivery of opti- mal FUS treatment. Ultrasound had been shown to help locate the ac- tual in situ focal point of FUS, and therefore, it is possible to compensate for navigational error due to beam distortion by the heterogeneous hu- man body. However, ultrasound still cannot assist in determining correct ultrasound dosage in a realistic clinical setting. Microbubbles has been investigated as a biocompatible, internal “probe” to convert a local par- ameter to an echo characteristic that can be measured externally. We accessed the main challenge is that the multiple acoustic parameters are not easily isolated from the multiple measureable characteristics of the echo signals (such as frequency shifts and harmonic component). In order to isolate the multiple factors (such as attenuation and perfusion Fig. 1 (abstract O65). The experimental setup rate) contributing to measurable echo characteristics, we sought to ex- ploit the highly specific behaviors of microbubble destruction when ex- posed to intense ultrasound. This paper reports a feasibility study of a pre-treatment scheme to determine effective attenuation and other rele- vant parameters and subsequently compensate for them during the ac- tual therapeutic procedure. METHODS A multi-channel transmission system and an array transducer was designed and built (Fig. 1). The current transmission system provides 128 physical channels that transmit arbitrary waveform with an analog bandwidth of 12 MHz at 4W sustained power or 50W peak power. The number of channels is scalable up to 1024. The transducer consists of a 125-element 2-D array operating at 2.1MHz. The complete system pro- duces a high resolution focal spot of 0.8x0.8x5 mm, steerable to +/- 8 mm in any direction. Highly flexible treatment plans can be implemented with successive focal points can be targeted at an extremely high rate. Thus arbitrary exposure fields can be achieved. A phantom containing Sonovue ® microbubbles are exposed to the treatment beam. The treat- ment focus was scanned to expose an ROI of 8 mm diameter. A programmable high frame rate scanner system (Verasonic Vantage-128, USA) was used to capture echo data at very high spatial and temporal resolutions. A custom imaging sequence produced good quality B-scan frames at a rate of 1 kHz, which are interleaved between FUS exposure at various burst lengths and amplitudes. The treatment beam is normal to Fig. 2 (abstract O65). Destruction curves of microbubble in the the imaging plane. Reconstructed echo image frames were analyzed to treatment ROI for different excitation voltage at a specific locate the treatment ROI automatically. The dynamics of the averaged burst length echo signal in the ROI as functions of treatment parameters was Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 45 of 122 RESULTS Oxy-NGV was developed and characterized to be rod shape with 400 nm length and 100 nm diameter. Through cell-free experi- ment, oxygen inside oxy-NGV can be released by applying 0.8 MPa ultrasound pulse, consequently increasing oxygen concentration in solution immediately. Oxygen releasing rate and amount can be spa- tiotemporally controlled by ultrasound intensity. Cytotoxicity test in MCF-7 and HeLa cell shows increased cell death rate in Oxy-NGV me- diated SDT group compared to pure PpIX one. The improvement of cell death rate is probably attributed to the increased singlet oxygen level as proven by increased SOSG fluorescence intensity. Meanwhile, both the cytotoxicity and singlet oxygen amount have positive cor- relation to the extracellular and intracellular oxygen level by chan- ging Oxy-NGV concentration. In hopxia condition, the therapeutic improvement is more significant than normal oxygen condition. CONCLUSIONS In summary, Oxy-NGV can efficiently deliver oxygen in a presicely controlled manner and consequently enhance SDT out- come through the mechanism of enhancing singlet oxygen produc- tion. NGV, as a novel stable nanometer size contrast agent and oxygen carrier, has great potential to enhance other oxygen medi- ated cancer therapy like radiotherapy, chemotherapy and photo- dynamic therapy. Fig. 3 (abstract O65). Extent of destruction measures at various incident intensities and burst lengths. Note that both axes are not linearly scaled O67 Synergistic ablation of tumours in vivo by high-intensity focused ultrasound and ethanol H. Murad, G. Halliburton, D. Luo, H. Yu, D. Khismatullin O66 Biomedical Engineering, Tulane University, New Orleans, Louisiana, USA Enhanced sonodynamic therapy using oxygen-rich nano gas Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O67 vesicle Y. Yang, X. Hou, L. Sun OBJECTIVES High-intensity focused ultrasound (HIFU) emerges as a Interdisciplinary Division of Biomedical Engineering, Hong Kong powerful technology for noninvasive or minimally invasive non- Polytechnic University, Hong Kong, China ionizing treatment of cancer, with recent FDA approval. HIFU de- Correspondence: Y. Yang posits a large amount of acoustic energy at the focal region within Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O66 the target tissue (i.e., tumor), causing tissue heating and necrosis, a process known as thermal ablation. Noninvasiveness is an important OBJECTIVES Sonodynamic therapy (SDT), based on the synergistic ef- advantage of HIFU over other thermal ablation methods, and our la- fect of low intensity ultrasound and sonosensitizer, is a potential boratory explores the ways for synergistic combination of HIFU with noninvasive approach for the treatment of cancers. Singlet oxygen, other therapeutic modalities to achieve the complete destruction of the major cytotoxic agent, is generated from dissolved oxygen within large and multifocal tumors. In this study we test the hypothesis that treatment region. However, because of the hypoxia microenviron- HIFU and percutaneous ethanol injection (PEI), a leading method for ment of solid tumor, oxygen deficiency restricts the singlet oxygen chemical ablation, have a synergistic effect on ablation of aggressive generation and consequently dramatically decrease therapeutic ef- liver and prostate cancers in vivo. fect. To locally increase the oxygen level may potentially improve the METHODS This in vivo study was performed using the xenograft treatment outcome. Here, we reported an oxygen enhanced SDT mouse models of human liver and prostate cancers. Hep3B human method utilizing oxygen-rich nano gas vesicle (Oxy-NGV). The objec- cancer cells and DU145 human prostate cancer cells (2.0×106) were tives of this study is firstly to develop oxygen-rich nano gas vesicle injected on flanks of athymic nude mice. Tumors were allowed to and evaluate the oxygen releasing mechanism, then to investigate grow to 8-10 mm size and then separated into the following treat- the improvement of the new oxygen-rich nano gas vesicle enhanced ment groups: HIFU alone, PEI (50%Etoh, 50 μl) alone, PEI+HIFU sonodynamic therapy as well as the underlying mechanism. If this (50%Etoh, 50 μl), and sham. Tumor sizes were measured by caliper Oxy-NGV based oxygen delivery strategy proven to be efficient, it will every day and a veterinary diagnostic ultrasound system was used have great potential to extend applications to other oxygen based pre-treatment, 5 days, and 12 days’ post- treatment. Tumor volumes cancer therapy like radiotherapy, chemotherapy and photodynamic were calculated from the ellipsoid formula V=πabc/6, where a, b, c therapy. are tumor sizes in three orthogonal directions. Tumors were surgi- METHODS Oxy-NGV was developed based on the nano gas vesicle cally removed and fixed using 10% formaldehyde solution. Samples (NGV) produced by cyanobacteria. It was then characterized regard- were sent for H&E staining with a single blinded pathologist, and ing basic nanoparticle’s property and more importantly the oxygen live/dead percentages of tumor cross sections were determined at 5 releasing efficiency through being broken by ultrasound pulse (0.8 and 12 days post treatment. Cryogenic-Scanning Electron Microscopy MPa). The therapeutic effect was evaluated by in vitro cytotoxicity by (Cryo-SEM) was also used to capture membrane disruption post HIFU ultrasound treatment for 5 mins with the intensity of 5W/cm after +PEI exposure on DU145 prostate cancer cells. MCF-7 and HeLa tumor cells were incubated with sonosensitizer Pro- RESULTS Tumor growth is significantly reduced or completely elimi- toporphyrin IX (PpIX) and Oxy-NGV. Then the singlet oxygen level, as nated in tumors treated with HIFU in combination with PEI (Fig. 1). the major cytotoxic agent, was imaged using Singlet Oxygen Sensor Tumors treated with HIFU alone showed a decrease in tumor size at 5 Green (SOSG) in both cell-free model and intracellular scenario. days, then rebounding to similar sizes as the sham. PEI alone tumors Meanwhile the oxygen level was also tested by dissolved oxygen showed no significant reduction in size and continued to grow. Hist- meter compared with conventional SDT method. These studies were ology shows largest necrotic tissue area in tumors treated with PEI and repeated in both normal oxygen level and hypoxia condition. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 46 of 122 HIFU at 5 and 12 days’ post treatment. Cryo-SEM images show large was delivered via either H105, an unfocused 52 mm disc transducer, macropore (>1-2 micrometer) formation in cells treated with PEI+HIFU. or H185D, a 49 mm disc transducer with three cylindrical lenses, each CONCLUSIONS The combination of HIFU and PEI shows a synergistic focused to a depth of 20 mm from the exit plane. Our US pulse dura- effect on tumor destruction at very low concentration of ethanol and tions spanned 19 μs–22 ms, with PNPs spanning 0.6–6.9 MPa. 24 lower acoustic power. This combination may become an effective hours after surgery, pigs were sacrificed to harvest treated and con- minimally invasive treatment option that’s safer for patients with liver trol liver lobes. After sectioning, spatially- mapped samples were ana- or prostate cancer. Utilizing these two FDA approved treatment in lyzed for luciferase expression. combination could lead to a disruptive and translational method of RESULTS Our ongoing experiments have added further support for a tumor treatment. species-generalized model that increasing pulse duration enables lower PNP for effective UMGD. Within a paired study, increasing pulse duration used with H185D from 19 μs to 200 μs at a constant 6.9 MPa PNP yielded up to 17-fold increases in sampled luciferase ex- pression. In a repeated H185D study, we have also shown an increase in expression using lower-pressures and longer pulse durations (200 μs, 4.5 MPa and 2 ms, 2.7 MPa groups vs. the 19 μs, 6.9 MPa group). Nevertheless, pulse durations above 2 ms paired with lower pressure did not appear to further enhance expression. Despite these expres- sion increases, ALT and AST values were comparable or lower in groups using longer pulse durations compared with groups using a 19 μs pulse duration. Experiments using H105 yielded a similar trend. Relative to an 18 μs, 2.7 MPa condition, we found increased expres- sion in groups with conditions of 1 ms, 1.2 MPa; 4 ms, 0.8 MPa; as Fig. 1 (abstract O67). LEFT: Prostate tumor (8x9mm) before PEI well as 22 ms, 0.5 MPa. Our 4 ms, 0.8 MPa group resulted in elevated +HIFU ablation, RIGHT: Tumor was eliminated at 2 weeks with AST levels, however changes in ALT and AST values of all other groups no re-occurrence were nominal. From these experiments, we analyzed several spatial considerations between the two transducer designs. Of the two, the unfocused H105 treats a larger tissue volume simultaneously, but in ex- change it generates lower PNPs at maximum operational power. Con- versely, the cylindrically-focused H185D treats a condensed tissue volume which allows higher peak pressures at its operational limit. H105 outperformed H185D when comparing conditions of equivalent pulse duration and PNP, a reasonable result given H105’slargervolume. When instead comparing the two transducers in terms of average in- tensity across their entire active area, we found no obvious discrepancy in expression. However, when using H185D in its higher range of PNPs (4.5-6.9), individual tissue segments were observed to have higher max- imum expression than the maximum values sampled using H105 at its limit PNP of 2.7 MPa. We are currently investigating whether this stems Fig. 2 (abstract O67). LEFT: Liver tumor (7x10mm) before PEI from the higher maximum PNP generated by H185D or whether using +HIFU ablation, RIGHT: Tumor was eliminated at 2 weeks with the transducer at those higher powers recruits greater UMGD efficacy no reoccurrence from weakly-focused areas. CONCLUSIONS By manipulating US pulse durations, our group achieved increased gene expression following UMGD in pigs. Such O68 tuning has also allowed comparable expression at decreased PNPs, Multivariate ultrasound signal manipulation for effective gene circumventing voltage limitations of piezo-materials. Since attenu- delivery to pig livers ation impedes high PNPs in transcutaneous UMGD applications, 1 1 1 1 2 J. Harrang , S. Song , M. Kajimoto , J. Chen , R. Fu1, K. Morrison , these results have promising implications for that modality. Our re- 2 1,3 G. W. Keilman , C. H. Miao sults demonstrate the advancement of efficient gene transfer in large Center for Immunity and Immunotherapies, Seattle Children's Research, animal models and we have also begun to elucidate the spatial re- Seattle, Washington, USA; Sonic Concepts Inc., Bothell, Washington, quirements for effective UMGD. USA; Pediatrics, University of Washington, Seattle, Washington, USA Correspondence: J. Harrang Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O68 OBJECTIVES Over several years our group has increasingly shown O69 that ultrasound-mediated gene delivery (UMGD) can be enhanced by Destruction of staphylococcus aureus biofilms on surgical mesh selecting favorable ultrasound (US) parameters. In particular, increas- T. A. Bigelow, H. Wu, C. Thomas ing pulse duration lowers the peak negative pressure (PNP) required Iowa State University, Ames, Iowa, USA for UMGD in mouse and cell models. This effect allows selection of Correspondence: T. A. Bigelow conditions with minimal associated tissue damage and enables tun- Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O69 ing to maximize the capabilities of piezo-materials. We have now fur- ther investigated scalability of this finding in pigs. In parallel, we OBJECTIVES It may be possible to significantly reduce the ~35,000 examined several spatial effects which represent important consider- surgeries each year needed to replace infected surgical meshes fol- ations for eventual clinical application of this novel technology. lowing abdominal hernia repair. The use of a surgical mesh has be- METHODS We have established an open surgery protocol used to ex- come standard medical practice. However, mesh infection is a plore optimal US parameters for efficient UMGD in pigs. First, the significant complication with incidence rates ranging from 1% to liver of each pig was exposed via a midline incision. Next, using con- over 10%. Mesh infections require reoperation for mesh removal trast US to confirm placement and perfusion, we catheterized a spe- ~70% of the time resulting in the potential for hernia reoccurrence cific branch of the portal vein. Just prior to therapeutic US exposure, and the need for additional operations. Therefore, there is a critical the inferior vena cava was temporarily occluded. Then US exposure need to develop new methods to noninvasively treat mesh infections and infusion of a solution containing pGL4 plasmid and phospholipid without removing the mesh. Our goal is to develop ultrasound microbubbles (MBs) were initiated simultaneously. Therapeutic US cavitation-based histotripsy to treat infections on surgical mesh. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 47 of 122 METHODS S. aureus biofilms were grown on 1 cm square surgical mesh samples for 3 days. The samples were then rinsed with phos- phate buffered saline prior to being inserted into Aquaflex Ultrasound Gel Pad Standoffs (Parker Laboratories Inc., Fairfield, NJ). The gel pads are bacteriostatic to minimize bacteria growth and have approximately the same mechanical properties as abdominal muscle. Mechanical properties have a significant impact on cavitation treatment (such as those used in our study), and therefore it is important they be relatively similar to real biological tissue. Each gel pad was cut in half allowing for two experiments per gel pad, and a slit was then cut in the gel pad to allow the insertion of the infected mesh sample. The focus of a spherically focused transducer (1.1 MHz, 12.9 cm focal length, 12.7 cm diameter) was then aligned on the mesh samples using a low-power signal from a pulser-reciever (Panametrics 5900, Olympus Corporation, Tokyo, Japan). Once aligned, the mesh samples were exposed to either a sham expsosure or histotripsy pulses (compressional pressure of 155 MPa, rarefactional pressure of 17 MPa) with tone burst durations of 3, 5, or 10 cycles at a pulse repetition frequency of 333 Hz for a duration of 15 seconds per exposure location with 5 repetitions per exposure group including the sham. The entire mesh was treated by scanning the focal spot in a raster pattern over the mesh using a step size of 750 Fig. 2 (abract O69). The number of S. aureus CFUs left on the gel μm. After treatment, the number of colony forming units (CFUs) on the pad following the ultrasound exposures mesh and the surrounding gel was independently determined. RESULTS For the mesh samples (Fig. 1), sham exposures have statisti- cally significantly more colony forming units than each of the treatment O70 groups. The absence of a bar corresponds to when no CFUs were A special retinal ganglion cell responses to low-frequency focused found. If we compare the means, we see a reduction of 2.00-log10 for ultrasound stimulation the 3-cycle treatment, 3.25-log10 reduction for the 5-cycle treatment, H. Zhao, Q. Jiang, G. Li, M. Su, H. Zheng, W. Qiu and a 3.23-log10 reduction for the 10-cycle treatment. However, the Shenzhen Institutes of Advanced Technology, Chinese Academy of differences between the 3-cycle treatment and 5- cycle/10-cycle treat- Sciences, Shenzhen, China ments are not statistically significant once a Bonferroni correction has Correspondence: H. Zhao been applied. More observations are needed to determine if the 3- Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O70 cycle exposures are significantly higher. The results from the gel pad (Fig. 2) showed no significant change in the number of CFUs for re- OBJECTIVES Acoustic retinal prosthesis has been put forward using leased bacteria from the biofilm for any of the treatments. However, high-frequency US with noninvasive and high- resolution advantages. the high variance in the numbers of CFUs for the gel samples means But its application is limited by fabrication, energy consumption and that while these numbers are not significantly different, we cannot mismatching to human eyeball’s anteroposterior axis. As our previous claim that additional bacteria are not being released by the ultrasound study demonstrated, the spatial resolution of low-frequency focused exposures. Given that approximately 40% of the bacteria in the sham US (LFUS) can be improved by decreasing applied acoustic intensity. experiments were on the mesh, we would need to show that the bac- Thus we prefer the acoustic retinal prosthesis using LFUS and are in- teria in the gel are not statistically greater than this to make this claim. terested in the electrophysiological properties of retinal ganglion cell This would require more observations per treatment group. (RGC) responses. This study has inspected the characteristics of one CONCLUSIONS The ultrasound histotripsy treatments are effectively special type of RGCs’ responses to LFUS in comparison with their destroying most of the biofilm on the infected surgical mesh. We ex- light responses, and examined the response changes in the presence pect that further optimization of the exposure parameters will further of ON pathway blocker. enhance destruction of the biofilm. METHODS A 2.25 MHz focused US transducer (D=0.75 in., SF=2.0 in.) was used to stimulate retina which was cultured in a multi-electrode array system (MEA2100, MCS, Fig. 1a). The acoustic property was evaluated by hydrophone (UMS3, Precision acoustics). US stimulation was modulated at pulsed mode (Fig. 1b). Light stimulation was mod- ulated in the same mode to give a uniform field flashes. The electro- physiological data collected from MEA was detected for neural spikes and sorted by Plexon Offline Sorter. Only channels recording single- cell activities were adopted for subsequent analysis. Peri-stimulus time histograms and raster plots were plotted for each RGC using Spike 2. RESULTS LFUS can activate RGC and generate sustained excitation (Fig. 1c). A comparison between light and US- induced responses in the same RGC shows differences in response temporal pattern and polarity (Fig. 1d). Light produces sustained excitation at stimulus onset and prolonged inhibition at offset, which means the recorded RGC is ON-centered. But US elicits a transient peak of excitation followed by delayed sustained excitation at stimulus onset, and transient excitation at offset. 21 investigated RGCs show the same differences. A typical result of ON pathway block- ing shows the delayed sustained excitation is eliminated by 100 μM L-AP4 (Fig. 1e). It is possible that this sustained excitation is generated from photoreceptors which transmission to bipolar Fig. 1 (abract O69). The number of S. aureus CFUs left on the mesh cells is blocked. The exemption of ON- and OFF-transient excita- following the ultrasound exposures. The absence of a bar indicates tion are probably because US directly activates interneurons or no CFUs remained on the mesh RGCs in retinal neural circuits. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 48 of 122 CONCLUSIONS We have investigated the neurophysiological properties of RGC responses to LFUS. We have discovered some new temporal re- sponse patterns of RGCs that haven’t been reported previously, including the characteristic dual-peak response patterns in ON-sustained RGCs and transient RGCs. In addition, we have found that US can modulate the temporal-spatial characteristics of RGC firing activities, which suggests that US can encode the information transmitted by RGCs. These results of our study will provide an important foun- dation for the development of ARP. Fig. 1 (abstract O71). See text for description Fig. 1 (abstract O70). See text for description O71 Guided longer pulses from a diagnostic ultrasound and intraclot microbubbles enhanced catheter directed thrombolysis 1 1 1 1 1 2 S. Gao , Q. Zhu , X. Dong , Z. Chen , Z. Liu , F. Xie Department of Ultrasound, Xinqiao Hospital, Third Military Medical University, Chongqing, China; Internal Medicine Fig. 2 (abstract O71). See text for description Cardiology, University of Nebraska Medical Center, Omaha, Nebraska, USA Correspondence: S. Gao O72 Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O71 Effects of f-number on the histotripsy intrinsic threshold and cavitation bubble cloud behaviour OBJECTIVES Insufficiency of MB in and around the vessel- E. Vlaisavljevich, T. Gerhardson, T. Hall, Z. Xu obstructing thrombi significantly reduces the effectiveness of University of Michigan, Ann Arbor, Michigan, USA ultrasound assisted thrombolysis (UT). Combined with intraclot in- Correspondence: E. Vlaisavljevich fusion of MB, guided longer pulses ultrasound from a diagnostic Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O72 transducer should be able to improve catheter directed thromb- olysis (CDT) procedure. OBJECTIVES Histotripsy is an ultrasonic ablation method that frac- METHODS In a thrombo-embolised rabbit IVC model, parallel to tionates soft tissue through the precise control of acoustic cavitation. catheter directed rt-PA thrombolysis procedure, guided moderate MI Previous work has demonstrated that a cavitation cloud can be longer pulses from a modified diagnostic ultrasound transducer formed by a single acoustic pulse with one high amplitude negative combined with an intraclot infusion of MB were aplied to facilitate cycle when the negative pressure exceeds an intrinsic threshold of CDT. The thrombolysis efficacy score, pre and post-treatment plasma ~25-30 MPa. Although previous work has provided significant insight concentration level of D-Dimer, a product of fibrinolysis, were into the process of intrinsic threshold histotripsy, the majority of acquired and compared in the four groups (CDT+UT, CDT alone, UT these studies have used highly focused (i.e. f-number<0.6) transduc- alone, & control). ers. In this study, we investigate the effects of f- number on the his- RESULTS The higher thrombolysis efficacy score (Fig. 1) and consist- totripsy intrinsic threshold and cavitation bubble cloud behavior, ent elevated post-treatment plasma concentration level of D-Dimer which is essential to the development of histotripsy for different clin- (Fig. 2), both indicated a superior of CDT + UT over CDT/UT alone. ical applications. There were no evidences of thrombo-embolism or local thrombus METHODS The effects of f-number on the histotripsy intrinsic thresh- formation in the cardiac-pulmonary vessels. old and cavitation bubble cloud behavior were investigated using a CONCLUSIONS Combined with intraclot infusion of MB, guided lon- 235-element 500 kHz array transducer, with the effective f-number of ger pulses from a diagnostic transducer was able to improve catheter the transducer varied from 0.51 to 0.89 by changing the active ele- directed thrombolysis procedure. This strategy has a possibility to ments in the array. Ultrasound pulses of 1-2 acoustic cycles were ap- achieve earlier clot removal, lower dosage of thrombolytic agent ad- plied to tissue mimicking phantoms, and the resulting cavitation ministrated, and in consequence lower incidence of thrombolysis re- activity was detected and characterized by passive cavitation detec- lated side effects. tion and high-speed photography (Phantom V210, Vision Research). Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 49 of 122 The intrinsic threshold at each f-number was defined as the pressure generated atdifferent microbubble-cell distances remained to be eluci- at which the probability of generating cavitation was 0.5. Optical im- dated. The goal of this study was to investigate how the microbubble- ages were further analyzed to determine the effect f-number on bub- cell distances influence the cell dynamics, such as the cytoskeleton ar- ble cloud characteristics including the bubble cloud dimensions, the rangement and the cellular membrane permeability, and the experimen- “bubble density” within the cloud, and individual bubble size. Finally, tal observations were compared with theoretical simulations. the effect of f-number on histotripsy fractionation efficiency was in- METHODS The in situ cellular responses (e.g., cytoskeleton arrange- vestigated by applying histotripsy to tissue phantoms embedded ment and intracellular delivery) to microbubble-mediated sonopora- with a layer of red blood cells, with the resulting fractionation visual- tion process generated withdifferent microbubble-cell distances were ized using optical imaging. systemically assessed based on an integrated system combining ultrasound exposure apparatus with real-time fluorescencemicro- RESULTS The intrinsic threshold did not significantly change with f- scope imaging. The microstreaming and shear stress generated by number, with the threshold remaining ~27-30 MPa for all conditions an oscillating microbubble was simulated based on an encapsulated (Fig. 1A). The predictability of intrinsic threshold histotripsy was fur- microbubble dynamic modelwith considering nonlinear rheological ther demonstrated by experiments showing close agreement be- effects of both shell elasticity and viscosity. tween the predicted and experimentally measured bubble cloud RESULTS The results show that: (1) as the microbubbles get closer to dimensions for all f-numbers. Quantifying the size of individual bub- cells, the disassembly of the cytoskeleton will accelerate; (2) as the bles formed directly above the intrinsic threshold at different f- num- microbubbles get closer tocells, the permeability of cell membrane bers showed no significant change in bubble size (~300 μm) with f- will have significant improvement; (3) the maximum microstreaming- number. Comparing bubble clouds at different f-numbers showed a induced shear stress on the cell membrane increasesrapidly with re- significant reduction in the “bubble density” with increasing f- ducing the bubble-cell distance. number, ranging from 39.6±3.8 bubbles/mm for an f-number of CONCLUSIONS In summary, by performing live cell fluorescent im- 0.51 to 1.5±0.3 bubbles/mm for an f-number of 0.89 (Fig. 1B). Finally, aging over the process of sonoporation, the accelerating cytoskel- experiments comparing the efficiency of histotripsy ablation demon- eton disassembly and improvedmembrane permeabilization could strated a significant increase in treatment efficiency for lower f- num- result from microbubble-mediated sonoporation with reducing bub- bers. For example, the number of pulses required to fractionate 50% ble-cell distance. The shear stresses resulting from microstreaming- of the treatment zone increased from 14.5±9.5 pulses for an f- generated nearby pulsating bubbles with different bubble-cell number of 0.51 to 209.3±17.4 pulses for an f-number of 0.89, corre- distances were simulated theoretically, which could provide a better sponding to a >14-fold increase in treatment efficiency (Fig. 1C). explanation for observed phenomena. The result suggests that in CONCLUSIONS The results of this study demonstrate that the order to achieve more efficient sonoporation effect in therapeutic ap- histotripsy intrinsic threshold does not significantly change with f- plications, it is better to find and optimal bubble-cell distance number. In addition, results show that histotripsy fractionation andmanipulate the microbubble location more rationally under cer- efficiency decreases at higher f-numbers due to a decrease in the tain conditions. “bubble density” within the bubble cloud. Overall, this study provides significant insight into the effects of f-number on intrinsic threshold histotripsy that will help to guide the design of histotripsy transduc- ers for specific clinical applications. O74 A pilot study of histotripsy using 1.1/2.2 MHz dual-frequency superimposition focused ultrasound pulses Y. Li, R. Wang, M. Lu, W. Huang, F. Ma, L. Zhang, M. Wan The Key Laboratory of Biomedical Information Engineering of Ministry of Education, Department of Biomedical Engineering, School of Life Science and Technology, Xi’an Jiaotong University, Xi’an, China Correspondence: Y. Li Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O74 OBJECTIVES Kidney cancer is a severe disease which can be treated using a non-invasive, controllable and focused ultrasound surgery method, termed as histotripsy. However, the time of le- sion formation following a single frequency histotripsy is long, so the dual-frequency mode histotripsy using second-harmonic superimposition hundred- microsecond pulses to reduce the le- sion time is very necessary. The aim of this research is to explore the feasibility of enhancing histotripsy by increasing effective Fig. 1 (abstract O72). See text for description cavitation and efficient boiling. METHODS By controlling the ratio of dual-frequency acoustic powers, the superimposition of two frequency pressures results in 9 split foci O73 along beam axial within confocal region, and the maximal peak in- Sonoporation-induced intracellular delivery and cytoskeleton tensity of split focus can reach about 2 times the sum of two fre- disassembly with different microbubble-cell distances quency intensities, indicating strong wave interference. Meanwhile, Maochen Wang, Chenliang Cai, Pengfei Fan, Dongxin Yang, Zhiyang Jin, the cavitation threshold lowers and the nonlinear effect strengthens Juan Tu, Xiasheng Guo, Dong Zhang by the dual-frequency superimposition mode. The efficient boiling School of Physics, Nanjing University, Nanjing, Jiangsu, China mechanism becomes dominant in histotripsy. The experiments im- Correspondence: Maochen Wang plemented in polyacrylamide phantoms with bovine serum albumin Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O73 and in porcine kidney ex vivo. The simplified two-stage hundred- microsecond pulsing scheme was used. The lesion formation process OBJECTIVES Sonoporation mediated by ultrasound-driven microbub- in the BSA phantom was monitored by high-speed photography and bles is being extensively studied as a promising technology to facilitate passive cavitation detection. gene/drug delivery to cells, and microstreaming generated by pulsating RESULTS The lesion inception time in gel-phantom was about 0.3s, microbubble near the cell membrane is regarded as one of the most im- which was 6 times shorter than that in single frequency mode (Fig. 1a). portant mechanisms in the sonoporation process. Thepresence of micro- The enhanced pressure, the lowered cavitation threshold and the bubbles is believed to help enhance ultrasound-induced bioeffects, but strengthened nonlinear effect by dual-frequency superimposition re- an in-depth understanding of cellular responses to sonoporation sulted in the lesion inception time decreased and boiling bubbles Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 50 of 122 emerged frequently in the axial split foci during each lesion formation, indicating enhanced histotripsy (Fig. 1b). The interaction of boiling bub- bles between split foci also had contributed to the lesion formation and the lesion size increased both radial and axial directions. The final lesion exhibited a long tear shape with smooth border (Fig. 1c). The le- sion generated in ex vivo porcine kidney was shown in Fig. 2, and the voids appeared with no marked thermally coagulated component remaining after the homogenate had been removed. The root mean square (RMS) amplitude of the broadband noise from filtered passive cavitation detection (PCD) data revealed the strong inertial-cavitation activities, and the increase of RMS demonstrated that the boiling bub- bles arose (Fig. 3a). Meanwhile, the inertial cavitation effect transferred to a higher band increment, which was beneficial to the frequency ab- sorption efficiency (Fig. 3b). CONCLUSIONS This study demonstrated the feasibility of enhancing Fig. 2 (abstract O74). Representative gross morphology of the dual- histotripsy by increasing acoustic intensity, lowering cavitation thresh- frequency-superimposition histotripsy lesions induced in ex vivo porcine old and strengthening nonlinear effect using dual-frequency superim- kidney with dimensions of approximately 6.0×2.8 mm (axial×lateral) position focused ultrasound pulses. The increase of effective cavitation and boiling dominated in the lesion formation. The split foci occur by controlling the ratio of dual-frequency acoustic powers. The maximal peak intensity of split focus can reach about 2 times the sum of two frequency intensities, indicating strong wave interference. The lesion in- ception time decreased, compared with single frequency mode. The final lesion exhibited a long tear shape with smooth border. The root mean square amplitude of the filtered passive cavitation detection data revealed the strong inertial-cavitation activities, and the increase of RMS demonstrated the boiling bubbles arose. Fig. 3 (abstract O74). (a) RMS amplitude of the PCD signals using dual frequencies of 1.1/2.2 MHz on the phantom at different times. (b)Broadband noise in the frequency domain using dual frequencies of 1.1/2.2 MHz on the phantom O75 Ultrasound-guided high intensity focused ultrasound ablation for diffuse adenomyosis J. Chen, Y. Feng, W. Chen College of Biomedical Engineering, Chongqing Medical University, Chongqing, China Correspondence: J. Chen Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O75 This abstract is not included as it has already been published: Feng Y, Hu L, Chen W, Zhang R, Wang X, Chen J. Safety of ultrasound-guided high-intensity focused ultrasound ablation for dif- fuse adenomyosis: A retrospective cohort study. Ultrason Sonochem. 2017; 36: 139-145. Available from: http://www.sciencedirect.com/sci- ence/article/pii/S1350417716304096. O76 Preliminary results of synthetic aperture imaging using random phased array M. Zubair, R. J. Dickinson Bioengineering, Imperial College London, London, United Kingdom Correspondence: M. Zubair Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O76 Fig. 1 (abstract O74). (a) The lesion inception time in gel-phantom OBJECTIVES Randomized phased arrays have been used for generat- was about 0.3s. (b) Boiling bubbles with about 1mm diameter in ing and steering single focus and multiple foci with low levels of phantom acquired by high-speed photography. (c) The final lesion grating lobes due to the breakage of periodicity of the elements and exhibited a long tear shape with dimensions of approximately are considered as useful source of HIFU. However, the reliance of 8.2×1.6 mm (axial×lateral) HIFU on MRI for real time visualization of the targeted tissue is a Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 51 of 122 major constraint in its clinical use due to the high cost of MRI and its low temporal resolution. There is a need to study the imaging cap- abilities of a therapeutic random phased array transducer for guiding the treatment process. METHODS Dual mode ultrasound phased arrays would have the ad- vantage of using the same array for both therapy and imaging due to the inherent registration between imaging and therapeutic frames of reference. The random spherical array would have limited field of view due to the fact that the array is optimized for therapy only and has large, directive elements sparsely positioned on a spherical sur- face. Nevertheless, images obtained will be useful for directing ther- apy as they will be perfectly aligned with the therapy transducer. Since strong scattering objects in path of HIFU beam are also in path of imaging beam, such scattering objects can be detected in real time and the HIFU beam can be adjusted accordingly. In our HIFU system the elements are randomly distributed with inter-element spacing much more than the required half a wavelength for reduced side lobes (Hand et. al. 2009), thus the spatial resolution of this sys- tem is poor. However, we use synthetic aperture imaging technique which has the potential to improve the spatial resolution of the ran- Fig. 2 (abstract O76). Grayscale images of wire target array using dom phased array. The simulations were carried out in MATLAB. The STA imaging approach numerical results were performed for a 1 MHz 256-element random phased array, made by Acublate Ltd, London, UK. Simultaneous foci were generated in simulations as well as experimentally based on the theory described by Gavrilov and Hand (2000a) (Fig. 1). RESULTS For imaging, preliminary simulations of synthetic aperture imaging with a 1MHz 256 element random phased array are shown. In Fig. 2, grey scale image of a wire target array is shown, which is a composite of point spread functions (psfs) at different on- and off- axis positions. The axial spacing between two wires is 10 mm, whereas the lateral distance is 5mm (figure 30. The -6 dB full width half maximum of the focused psf is 1.6 mm. Sub-apertures are being used to image the field of interest, where each sub-aperture contains only those elements which contribute to the pixel being imaged. This not only improves the resolution by decreasing the side lobes level but also reduces the computation time. CONCLUSIONS It was observed that random phased arrays are cap- able of therapy as well as imaging for treatment guidance. The use of sub-apertures improves the resolution and reduce the computa- tion time, however, it also limits the imaging field of view Fig. 3 (abstract O76). Lateral cross section at a depth 130 mm O77 Image-guided blood-brain barrier opening with focused ultrasound and microbubbles in mice using passive microbubble imaging 1 1 1,2 M. T. Burgess , I. Apostolakis , E. Konofagou Biomedical Engineering, Columbia University, New York, New York, USA; Radiology, Columbia University, New York, New York, USA Correspondence: M. T. Burgess Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O77 OBJECTIVES Blood-brain barrier (BBB) opening with focused ultrasound (FUS) and microbubbles is a technique for targeted drug delivery to the brain and has led to the development of ultrasound-guided focused ultrasound (USgFUS) systems for appropriate treatment monitoring and guidance. Comprehensive detection of FUS-stimulated microbubble ac- Fig. 1 (abstract O76). Distribution of 4 simultaneous foci at a depth tivity is critical for successful implementation of these systems. Current of 130 mm techniques passively image microbubble- related acoustic emissions Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 52 of 122 with ultrasound arrays to spatially map the magnitude and location of OBJECTIVES Lipid-coated contrast agents have attracted much atten- microbubble activity throughout the focal volume of the FUS. In turn, tion for contrast ultrasound molecular imaging as well as targeted this provides information about the extent and location of BBB open- treatment, and the connection between enhanced contrast capability ing. Current passive cavitation imaging methods with linear arrays suf- and acoustic properties of the lipid-coated ultrasound contrast fer from poor axial image resolution due to the use of long FUS pulses agents need to be elucidated. and asynchronous transmit and receive sequences that degrades abso- METHODS In this study, microbubbles were fabricated using thin- lute time-of-flight information. The objective of this study was to pre- film hydration and mechanical agitation. Then, the morphology and serve absolute time-of-flight information by using short pulses of FUS, distribution of these microbubbles were investigated through trans- along with synchronous transmit and receive acquisition, for utilization mission electron microscopy (TEM) imaging and dynamic light scat- of image reconstruction similar to pulse- echo B-mode imaging. It is ex- tering (DLS) sizing technology. To demonstrate physical properties of pected that this new passive microbubble imaging (PMI) technique will microbubbles, inertial cavitation threshold and acoustic attenuation improve image resolution and offer a methodology capable of high measurements were carefully assessed and shell parameters were resolution mapping of BBB opening. further estimated. The imaging function of synthesized microbubbles METHODS PMI was carried out during BBB opening with FUS and was also compared with that of SonoVue microbubbles in vivo. microbubbles in a mouse model. An 18-MHz imaging array (L22-14v RESULTS The results showed that the synthesized microbubbles had LF, Verasonics, Inc.) and 1-MHz FUS transducer were submerged in a a spherical shape, a smooth, consistent membrane and uniform dis- tank of degassed water and placed approximately 2 cm above the tribution, with average diameter of 1.484 μm. Imaging studies mouse skull for transcranial application of FUS. The FUS transducer showed that while exhibiting comparable imaging ability, synthe- was aligned at an acute angle relative to the imaging array’s axis. A sized microbubbles had a longer circulation time and better stability research-based ultrasound system (Vantage 256, Verasonics, Inc.) was than SonoVue. Physical characterization showed that compared with used to operate the imaging array in a passive mode and SonoVue, synthesized microbubbles with smaller interfacial tension and synchronize the FUS transmission with receive acquisition. Short dilatational viscosity, indicating less attenuation during the propagation pulses of FUS (2-3 cycles, 200-400 kPa) at a pulse rate of 500-5000 Hz of sound wave. In comparison with SonoVue, IC threshold of the micro- were used to insonify intravenously injected microbubbles as they bubbles are prominently higher in both concentration ranges, explain- flowed through the mouse brain microvasculature. In this scenario, ing better stability of the microbubbles. Present study built a bridge the FUS pulses were used for the dual purpose of imaging and ther- between physical properties and in vivo imaging performance of syn- apy (BBB opening). Receive acquisition frames were recorded for thesized microbubbles, which could provide guidance to the design each FUS transmit and the data was saved for off-line processing. and safe application of ultrasound contrast agents. PMI images were formed using a custom GPU-based image recon- CONCLUSIONS While the average grey scale of various organs all in- struction algorithm in MATLAB (The Mathworks, Inc.). The time delays creased following microbubbles application, tumor imaging showed for implementation of delay-and-sum beamforming were calculated that synthesized microbubbles stayed in the tumor area for longer to account for the propagation from the FUS transducer, to the period of time and has a longer enhancing time. The long-circulating mouse brain, and back to the imaging array. Blocks of PMI frames behavior showed that synthesized microbubbles may be better suited were further processed using spatiotemporal filtering to isolate for tumor imaging and therapeutic application in drug/gene delivery. microbubble emissions. PMI was compared with post-FUS dynamic Furthermore, in-vivo and tumor imaging performance of synthesized contrast enhanced-magnetic resonance imaging (DCE-MRI) to correl- bubbles was well explained by acoustic property measurements and ate microbubble activity with BBB opening. shell elastic and viscous parameters. Possible correlation between phys- RESULTS PMI overcomes the poor axial resolution concerns of previ- ical/acoustical properties and in-vivo/tumor imaging performance was ous passive imaging methods and provides detailed maps of micro- revealed in this study, and could be of help in future design and prac- bubble activity throughout the focal volume of the FUS transducer tical application of ultrasound contrast agents. during the BBB opening treatment. PMI is able to provide the de- tailed structure of the brain microvasculature in manner similar to power Doppler imaging, although only within the focal volume of O79 the FUS. Simplistically, PMI reveals the heterogeneous distribution of Pulse inversion based broadband subharmonic cavitation imaging microbubble activity within focal area that can be used to predict for monitoring high intensity focused ultrasound therapy where BBB opening is occurring. Indeed, post-FUS DCE-MRI images H. Zhong, X. Ma, M. Wan correlated with the distribution of microbubble activity by showing Biomedical Engineering, Xi'an Jiao Tong University, Xi'an, China contrast leakage into the brain in areas of microbubble activity. In Correspondence: H. Zhong addition to the detailed microbubble maps, PMI provides important Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O79 temporal information about the microbubble infusion kinetics and persistence within the focal volume of the FUS. OBJECTIVES This paper proposed a cavitation imaging method by CONCLUSIONS PMI can address a key technological limitation of extracting broadband subharmonic based on pulse inversion tech- poor axial image resolution with current passive cavitation imaging nique during high intensity focused ultrasound (HIFU) therapy to ob- techniques and provide a method capable of monitoring BBB open- tain monitoring images with high cavitation-to-tissue ratio (CTR), ing with FUS and microbubbles. Limitations of this new technique high cavitation detection sensitivity and high resolution. will be discussed along with its similarities and differences with exist- METHODS HIFU delivery is briefly interrupted with the HIFU com- ing methodologies. This new USgFUS modality is not only a promis- pletely off for transmission of a pair of inverted pulses (4.6 MHz) and ing technique to be used for treatment of CNS diseases, but also for acquisition of the backscattered signals from tissue samples during oncological, cardiovascular, and other medical conditions that utilize HIFU treatment. After summing the echoes of positive and negative FUS in combination with microbubbles. pulses, the subharmonic filters are designed with the center fre- quency of 2.3 MHz and the bandwidth of 20%, 60%, 80%, 100% and 140% to obtain the cavitation images. For comparison, the second O78 harmonic images were also obtained. A threshold is used to assess Long-circulating behaviour and acoustic characterization of the cavitation detection sensitivity of different cavitation images. lipid-coated microbubbles Cavitation bubble areas could be calculated by counting the number Y. Yang, H. Li, X. Guo, D. Zhang, J. Tu of pixels whose values are greater than the threshold. The larger is Nanjing University, Nanjing, Jiangsu, China the cavitation bubble area, the higher is the cavitation detection sen- Correspondence: Y. Yang sitivity. Cavitation-to-tissue ratio (CTR) could be defined as the ratio Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O78 of the intensity value averaged in the regions of interest (ROI) of Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 53 of 122 HIFU treated area to that in the ROI of HIFU untreated area. We used CONCLUSIONS These results provide more complete framework for a the normalized −6 dB width of the autocorrelation function of the large body of experiments in multiple preparations across the field of enveloped RF signal to quantify the resolution of cavitation images. US neuromodulation, lending further support to the hypothesis that RESULTS The experiments with porcine muscle demonstrated that intramembrane cavitation is responsible for ultrasonic neuromodula- the cavitation images obtained by using 80%~100% bandwidth sub- tion. They could thus pave the way towards new CNS therapeutic harmonic filters have the greatest cavitation detection sensitivity. protocols, using the only method that currently allows targeted non- The cavitation bubble areas of broadband subharmonic images invasive neuromodulation with millimeter spatial resolution essen- (1.15~3.45 MHz, including 1/2, 1/3, 1/4 subharmonic components) tially anywhere in the brain. are 1.6~2.7 times of those of narrowband subharmonic images (2.07~2.53 MHz, including 1/2 subharmonic component) with the suitable threshold. The difference of the CTR values between the O81 broadband and narrowband subharmonic images is not very large, Non-invasive high frequency transcranial focusing with a single but the CTR values of both broadband and narrowband subharmonic element transducer: experimental validation of adaptative images are much greater than those of second harmonic images. It focusing through human skulls with a 3D printed acoustic lens 2,1 2 2 2 2 was found that the resolution of broadband subharmonic images G. Maimbourg , A. Houdouin , T. Deffieux , M. Tanter , J. Aubry 1 2 was improved up to about 2 times compared with narrowband sub- Université Paris Diderot, Paris, France; Institut Langevin, ESPCI Paris, harmonic images. CNRS UMR7587, INSERM U 979, Paris, France CONCLUSIONS The proposed broadband subharmonic cavitation im- Correspondence: G. Maimbourg aging method could obtain much higher CTR value than second har- Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O81 monic imaging method. And the method has higher cavitation detection sensitivity and resolution than normal narrowband subhar- OBJECTIVES Transcranial ultrasonic brain therapy requires a wave monic imaging method. front shaping device to compensate skull- induced aberrations. Up to now, this was done with a multi-elements probe: the phase of the signal emitted by each individual transducer is adjusted in order to O80 compensate for the delay caused by the crossing of the skull. Histor- A generalized theoretical framework for understanding ultrasonic ically, a growing number of elements was used (64 elements in 2000 neuromodulation mechanisms [1], 300 in 2003 [2], 1024 in 2012 [3]) to improve the focusing. We M. Plaksin, S. Shoham, E. Kimmel propose to radically change the paradigm by achieving adaptive Faculty of Biomedical Engineering & Russell Berrie Nanotechnology transcranial focusing with a single- element covered with a 3D Institute, Technion – Israel Institute of Technology, Haifa, Israel silicone acoustic lens of variable and controlled thickness [4]. Similar Correspondence: M. Plaksin lenses have been introduced in the past to perform single or Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O80 multiple focusing patterns in homogenous propagating media [4,5] but recent 3d printing and milling capabilities make tailor-made 3D OBJECTIVES Low intensity US can noninvasively suppress or excite lenses a feasible option for transcranial adaptive focusing. The char- central nervous system (CNS) activity. Recently, we introduced the acteristic length of variation in thickness for the acoustic lens is less Neuronal Intramembrane Cavitation Excitation (NICE) framework to than 1mm. A 6cm radius lens- corrected single-element is thus explain the observed aspects of ultrasonic neuromodulation, through equivalent to an 11,000 element transducer in terms of phase shap- intramembrane space-related capacitance variations leading to cell- ing capabilities. type-selective activation effects. Here we expanded the framework to METHODS The lens is made of silicone (Elite double 8, Zhermack dissect also the impact of acoustic radiation pressure (ARP) - induced Spa, Italy). The speed of sound is c =1000m/s in silicone and silicone membrane capacitance changes on neuronal activity, as well as to c =1485m/s in water. This difference can be used to modify the water explore the effect of cell environment effective viscosity on NICE wave phase by controlling the local thickness of the lens, and thus model-related neural response. correct skull-induced aberrations [6]. The study was conducted on three METHODS We analyzed the relevant experimental literature using human skulls. The skulls were harvested and cleaned at the Saints- two sources of ARP gradients responsible for membrane dynamics: 1) Pères Anatomy Institute (Paris Descartes University) for transcranial ARP caused by ultrasonic field inhomogeneity and 2) ARP caused by ultrasound focusing studies. A 3D simulation based on computed- acoustic impedance mismatch. In addition, live brain tissue ARP- tomography of the skulls was then performed to estimate the phase gradients-subjected areal strains were evaluated in a viscoelastic shift induced by the skull at surface of the transducer (single element, brain model. In the context of the NICE model dynamics, the modi- 59 mm radius of curvature, f-number of 1, operated at 914 kHz). The fied Rayleigh–Plesset-based intramembrane cavitation biomechanics thickness of the lens was then adjusted to compensate the shifts by was calculated with cell environment viscosity higher than water vis- casting the silicone in a 3D-printed mold. The acoustic-lens-covered cosity (water viscosity was used in our previous studies), to express transducer was then immersed in water (Fig. 1). Lastly, the skull was po- the exact biological properties of cells. By coupling these biomechan- sitioned in front of the transducer with a 3D printed holder. The quality ical models to biophysical membrane models we predict dynamical of the focusing through the skull was then assessed using a 3D scan of biophysical responses of artificial bilayer membranes, and of com- the pressure field with a needle hydrophone (HNA-0400, Onda Corp., mon neocortical single cell Hodgkin-Huxley type models. Sunnyvale, CA, USA). RESULTS The augmented viscosity conditions in the NICE model lead RESULTS Figure 2 shows the acoustic pressure obtained experimentally to parabolic relationship between US frequency and threshold inten- with skull A. Axial and transverse views are presented in the absence of sities for cortical pyramidal neuron stimulation, wherein below 1 MHz skull (left), through the skull without correction (center) and with the the dependence on US frequency is negligible (with complete agree- acoustic lens (right). The lens qualitatively restores the focusing. Table 1 ment to our previous studies). The new results were found to explain reports the quantitative outcomes for the three skulls noted A, B and C. and predict the experimental results of Ye et al., Ultrasound Med Biol. The lens increased the maximum acoustic intensity by 97 ± 56 %. Com- 2016 that also captured the same behavior in their mouse study. The pared to the reference in water with no skull in place, the mean -3dB emergence of ARP gradients-induced membrane capacitance variations volume of the focus increased by 472 ± 231 % when crossing the skull, associated with membrane area changes fully explain artificial mem- whereas it only increased by 86 ± 29 % with the lens- based correction. brane experimental results and may also be responsible for neural exci- CONCLUSIONS We demonstrated that in 3D printing can be used to tation at in- vitro experimental conditions. However, these capacitance create custom-made acoustic lenses for CT- based non invasive skull- changes were found to be highly unlikely sources for neural excitation aberrations correction, making it possible to compensate for skull at in-vivo experimental conditions, when considering the areal strains aberrations without a multi- element device. Due to its simplicity, the expected to form in brain tissue during normal sonication. acoustic lens is expected to be a cheaper and less cumbersome Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 54 of 122 solution than current multi-element probes. The ultrasonic absorp- Table 1 (abstract O81). Quantitative results for assesing the efficiency tion of the lens material remains to be optimized for thermal treat- of the acoustic lens. Outcomes were obtained for three skulls named A, ments, but the setup presented here could be of interest for low B and C energy treatments such as neuromodulation or blood brain barrier opening. Acknowledgements This work was supported by the Bettencourt Schueller Foundation and the "Agence Nationale de la Recherche" under the program “Future Investments” with the reference ANR-10-EQPX-15. O82 Correlation of the lesion size in histology and mr images of the References pig brain tissue by transcranial MR-guided focused ultrasound [1] Clement G et al, A hemisphere array for non-invasive ultrasound brain 1,2 3 4 5 4 1,4 D. Paeng ,Z. Xu , J. Snell , A. H. Quigg , M. Eames , C. Jin , therapy and surgery. Phys Med Biol, 2000 [2] Pernot M et al., High power 3 3 6 4 A. C. Everstine , J. P. Sheehan , B. S. Lopes , N. Kassell transcranial beam steering for ultrasonic brain therapy. Phys Med Biol, 2003 Ocean System Engineering, Jeju National University, Jeju, Jeju, Korea; [3] Jeanmonod D et al, Transcranial magnetic resonance imaging-guided Radiation Oncology, University of Virginia, Charlottesville, Virginia, USA; focused ultrasound: noninvasive central lateral thalamotomy for chronic Neurosurgery, University of Virginia, Charlottesville, Virginia, USA; neuropathic pain. Neurosurg Focus, 2012 4 5 Focused Ultrasound Foundation, Charlottesville, Virginia, USA; Medical [4] Fjield T et al, Low-profile lenses for ultrasound surgery. Phys Med Biol, 1999 School, Virginia Common University, Richmond, Virginia, USA; [5] Melde K et al, Holograms for acoustics. Nature, 2016 Pathology, University of Virginia, Charlottesville, Virginia, USA [6] Patent: FR 1556217, July 2015 Correspondence: D. Paeng Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O82 OBJECTIVES This study is to investigate the correlation of the lesion size in histology and MR images of the pig brain tissue, which was formed by a transcranial magnetic resonance-guided focused ultra- sound (tMRgFUS) system and observed in 3 days after sonication. The lesion was generated by relatively low temperature between 46 ~ 52°C for an appropriate time to reach a target thermal dose in CEM (cumulated equivalent minutes). The target thermal dose was below 200 CEM which was obtained by variation of pulse duration of the tMRgFUS system for constant target temperature through a closed-loop system based on MR thermometry. METHODS A tMRgFUS system (ExAblate 4000 Neuro 650 kHz system, InSightec, Israel) was used for this pig experiment. An MRI system Fig. 1 (abstract O81). (A) Experimental setup in the water tank (Discovery MR75-3.0T, GE Medical systems) was used for thermom- during the 3D-scan of the pressure field by the needle hydrophone. etry and pre- and post-imaging. A closed-loop control system was The skull is supported on the internally-designed holder by two implemented on a personal computer to control pulse duration of elastic bands. (B) an axial cut of the setup. The acoustic lens covers the tMRgFUS system at a certain acoustic power in order to maintain the surface of the transducer a target temperature based on the MR thermometry. Temperature distribution and accumulated thermal dose in the target brain tissue was calculated every 3.7 seconds. Sonication was stopped when a prescribed thermal dose was delivered to the targeted tissue. The proportionate and integral coefficients of the PI controller were found to be 5 and 1.5, respectively, from several phantom experi- ments and first acute pig experiment while the acoustic power was varied up to a few hundred watts. Twelve chronic pig experiments with craniectomy were conducted, and post MR images within 1 hour and at 3 days were taken after sonication. Brain tissue was har- vested in 3 days before euthanasia for histology. Four lesions were generated on both sides of the thalamus of each pig. Temperature in the pig brain tissue was estimated by rectal temperature for the MR thermometry baseline. This study was approved by the University of Virginia Institutional Animal Care and Use Committee (ACUC). RESULTS Among 12 chronic pigs, one pig had a problem with intro- duction of air bubbles during surgery procedure and another one had an ACUC issue, so that these data could not be used. Three other pigs had a lower rectal temperature below 32°C so that the thermal dose computation was not reliable, leading to exclusion of Fig. 2 (abstract O81). Acoustic pressure acquired by a needle the data analysis. No obvious lesions were observed in MR images hydrophone. From left to right: the reference focusing (no skull and taken in an hour of sonication for thermal dose less than 200 CEM. no lens), the focusing through a human skull without any correction Figure 1 shows the lesion diameters measured in MR T2-weighted and the focusing through a human skull with the custom-made axial images and histology in 3 days of sonication as a function of acoustic lens. The black crosses depict the position of the reference thermal dose in CEM. There are lesions in all MR images and hist- focusing ology over 77 CEM except 101 CEM where no lesion is shown in Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 55 of 122 histology while MR shows a lesion. The lesion diameters from MR im- the DMUA behind the skull sample (Fig. 1). In other experiments, skull ages and histology are mostly 3 ± 1mm for thermal dose over 77 samples were embedded in a tissue- mimicking phantom and posi- CEM. There are 6 lesions in MR T2-weighted axial image but no le- tioned at a distance of approximately 32-mm from the apex of the sion was found in histology, whose thermal doses were 23, 30, 46, DMUA (corresponding to the skull position during in vivo experiments.) 53, 69, 101 CEM. Below 75 CEM, no lesion was found except one in A thermocouple was inserted so that its junction was closest to the geo- 18 CEM in histology. One lesion on 18 CEM is shown in both MR metric focus. Two modes of imaging were used: 1) Synthetic-aperture image and histology, and the rectal temperature was low to 33.3°C. imaging, which provided larger field of view to guide the placement of Lesion diameter observed in histology (y) is highly correlated with the tFUS beam, and 2) Single-transmit focus (STF) imaging, which one in MR T2-weighted axial imaging (x), and their function and cor- allowed for the characterization of the tFUS beam interaction with the relation coefficient are y=0.90x and r =0.66, respectively. Even skull. SA images were used to place target point(s) and critical point(s) though the histology is sliced in coronal plane, the histology lesion used by the refocusing algorithm. Raw echo data received by each array diameter is highly correlated with MR axial diameter. element were used to form data matrices corresponding to the target CONCLUSIONS The lesions were formed in the pig brain tissue in 3 and critical point(s). These matrices were used to solve an optimal re- days of sonication by tMRgFUS for thermal dose over 77 CEM, except focusing problem based on the concept of propagation operators from one in histology for thermal dose of 101 CEM. The diameter of the le- the array elements to the target point(s). The refocusing algorithm ran sions was measured to be mostly 3 ± 1mm in both MR T2-weighted automatically once the user specified the target and critical point. It was axial images and histology, which is close to focal size. The diameter implemented on a software- defined ultrasound (SDUS) architecture that in histology is 10 % smaller than the one in MR T2-weighted axial im- allowed the real-time computation of the refocused excitation vector ages with correlation coefficient of 0.66. There are some differences and immediate download to the driver within milliseconds. To demon- in lesions between histology and MR images, and further systematic strate the improvement in focusing gain, we have analyzed the change researches are required for the reasons. in echogencity from the target when insonified using geometric and re- focused STF imaging beams. In addition, the temperature rise due to therapeutic tFUS beams with geometric focusing and optimal refocusing were directly measured. The heating rates were computed by taking the time derivative of the measured temperature profiles. RESULTS Figure 1 also shows STF images generated from uniform elem- ent excitation (geometric focusing) and refocused excitation vector. The refocused STF image exhibits increased echogenicity from the target (thermocouple) and reduced echogenicity from the skull, where a critical point was located near the medial suture line. This is more clearly shown by the line graphs in Figs. 2 and 3. To demonstrate the increased thera- peutic gain due to refocusing, the geometric focusing beam was used in therapeutic mode to produce heating at the thermocouple for 10 second duration followed by a refocused beam. Figure 4 shows the heating rates produced by the geometric focus and the refocused beam as estimated from the thermocouple measurements. It is quite clear that the refocus- ing gain was increased approximately by a factor of 3. Specifically, the heating rate due to the refocused beam was 3.636 C/sec while that of the geometrically focused beam was 1.179 C/sec. This result was typical, but variation in the refocusing gain were observed depending on which part of the skull was traversed by the tFUS beam. CONCLUSIONS The results shown above demonstrate the feasibility of Fig. 1 (abstract O82). Comparison of the diameter of MR using STF imaging data for characterizing the quality of the tFUS beams T2-weighted axial image with the one of histology as a function noninvasively. More importantly, they demonstrate the feasibility of re- of thermal dose in CEM focusing the beam to provide significant improvement in focusing gain as evidenced by the 3-fold increase in heating rate shown. The result shown here was typical of numerous experiments where the tFUS O83 beam traversed different regions within the skull with different distor- Real-time image-based transcranial refocusing of dual-mode ultrasound arrays: ex vivo results tion. In every case, however, a significant improvement in focusing gain was achieved. The results shows were obtained using a 1D array and H. Aldiabat, P. D. O'Brien, D. Liu, E. S. Ebbini the refocusing was achieved only in the axial lateral directions. A 2D Electrical Engineering, University of Minnesota, Minneapolis, Minnesota, array would have produced even higher refocusing gain by taking ad- USA Correspondence: H. Aldiabat vantage of the elevation direction. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O83 OBJECTIVES Transcranial focused ultrasound (tFUS) is gaining wider ac- ceptance in a range of therapeutic applications for the treatment of brain disorders. A major challenge towards widespread use of tFUS- based therapies stems from the complexity of the skull that could result in severe loss of focusing gain. Using extensive transskull hydrophone scans in rodent and human skulls, we have documented a range of tFUS beam distortions from mild shifting to complete splitting. These distortions could compromise the specificity of targeting brain circuitry and/or cause collateral damage at unintended locations. In this paper, we present first ex vivo results demonstrating the feasibility of image- based refocusing to regain the focusing gain. METHODS A 3.5-MHz dual-mode ultrasound array (DMUA) prototype (64 elements, concave with 40-mm radius of curvature) was used. The DMUA was used to image and generate therapeutic focused ultrasound beams through rodent skull samples ex vivo. In some experiments, the Fig. 1 (abstract O83). See text for description target was the nose of a cone positioned near the geometric focus of Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 56 of 122 correction) acquired using the array elements as receivers. Following aberration correction, hydrophone measurements of the focal pres- sure amplitudes and focal beam profiles were acquired, as were mea- surements of the focal pressure amplitudes as a function of the electronic steering distance through the skull. Lesioning experiments were carried out in red blood cell tissue phantoms to compare how the different aberration correction modalities affected lesioning. Fi- nally, volumetric treatments of large clots through the skull were conducted and treatment efficacy for each aberration correction case was evaluated as a function of clot volumes liquefied. RESULTS Aberration correction performed based on hydrophone Fig. 2 (abstract O83). See text for description measurements and backscatter methods increased the peak negative pressure through the skull at the focus by 90% and 60%, respect- ively, compared to the no aberration correction case. The electronic steering radius at which focal pressures above the nucleation thresh- old could be reached improved by more than 100% with aberration correction, increasing from 7 mm without aberration correction to over 15 mm with aberration correction. The cavitation clouds and le- sions generated in both aberration correction cases were equal in size and up to 30% larger than the no aberration correction case using the same input power to the ultrasound array due to increased focal pressure amplitudes. The increased pressure amplitudes and steering ranges in the aberration correction cases resulted in signifi- cantly improved transcranial clot liquefaction speeds compared to the no aberration correction case. CONCLUSIONS This study indicates that all measures of histotripsy applied transcranially benefit from aberration correction. While hydrophone aberration correction was seen to provide the greatest increases in focal pressure, steering range, and liquefaction rates, ab- erration correction applied based on backscatter methods was seen to perform comparably to the hydrophone case and provide signifi- cant improvements over the no aberration correction case. These re- sults indicate that backscatter aberration correction can offer a non- invasive method of improving the efficacy of transcranial histotripsy comparable to hydrophone based correction methods. Fig. 3 (abstract O83). See text for description O85 O84 accumulated thermal dose in guiding essential tremor treatment: A comparative study of transcranial histotripsy applied with and repeated sonications with peak temperatures below 55°C without aberration correction 1 1,2 3 3 1,2 Y. Huang , R. M. Jones , N. Lipsman , M. L. Schwartz , K. Hynynen J. R. Sukovich, T. Gerhardson, T. Hall, J. Macoskey, C. A. Cain, Z. Xu 1 2 Sunnybrook Research Institute, Toronto, Ontario, Canada; Department Biomedical Engineering, University of Michigan, Ann Arbor, Michigan, of Medical Biophysics, University of Toronto, Toronto, Ontario, Canada; USA Division of Neurosurgery, Sunnybrook Health Sciences Centre, Toronto, Correspondence: J. R. Sukovich Ontario, Canada Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O84 Correspondence: J. R. Sukovich Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O85 OBJECTIVES Histotripsy is an ultrasound therapy that generates cavi- tation bubble clouds to fractionate soft tissue using short duration, OBJECTIVES Magnetic resonance guided focused ultrasound has re- high-amplitude ultrasound pulses. Previous studies have shown that cently been approved by the FDA for the treatment of essential histotripsy is capable of producing lesions through the skullcap with tremor. During these treatments, peak temperatures between 55 and or without using aberration correction. Transcranial ultrasound ther- 60°C are generally desired to produce reliable lesions at the target in mal therapy cannot treat within 2 cm of the skullcap and cannot the thalamus. However, in some patients a peak temperature of 55°C treat large volumes in the brain due to skull heating, which makes it cannot not be reached, either because of their skull properties or unsuitable for brain tumor treatments. In comparison, histotripsy can due to a low tolerance to the pain associated with substantial skull be used to treat targets near the skullcap as well as large volume tar- heating. In these patients, repeated sonications with peak tempera- gets transcranially with minimal skull heating by using low duty cycle tures from 50 to 54°C were typically achievable. The accumulated ef- pulses (<0.1%). This study presents comparative analyses of the pres- fects of these sonications may also result in a sufficient thermal dose sure fields, cavitation clouds and lesions, and treatment rates of his- to produce a lesion. Therefore, establishing a proper threshold for totripsy applied transcranially with and without aberration correction. the accumulated thermal dose (ATD) is important for guiding these METHODS Histotripsy pulses were delivered through ex-vivo human repeated sonications with lower peak temperatures. In this study, skullcaps mounted centrally within a 500 kHz, 256-element histo- clinical treatments at our institution with maximum peak tempera- tripsy transducer with transmit-receive capable elements. Aberration tures below 55°C were retrospectively reviewed. ATD maps were cal- correction was applied in two ways, 1) based on hydrophone mea- culated offline with corrections for chemical-shift artifacts and were surements and 2) based on histotripsy pulse-backscatter measure- correlated to the lesion sizes observed at follow-up imaging to find ments (from bubbles generated transcranially without aberration Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 57 of 122 the best estimate for the ATD threshold in terms of cumulative artery occlusion (dMCAO), and used both behavior (neurological se- equivalent minutes at 43°C (CEM ). verity score) and histological measures were used to assess the neu- METHODS From Jul 2015 to Jan 2017, 28 patients with medication- roprotection effects of TUS on ischemic stroke rats. In the second refractory essential tremor have been treated with a clinical focused study, we applied TUS stimulation as a preconditioning to rats before ultrasound system (ExAblate Neuro, 650 kHz central frequency, ischemia induced by photothrombosis, and both cerebral blood flow InSightec, Tirat Carmel, Israel) and a 3 T MR scanner (MR750, GE and histological measures was used to assess the neuroprotection ef- Healthcare, Milwaukee, WI, USA). Among these, 5 patients were fects of TUS preconditioning on ischemic stroke injury. treated with a maximum peak temperature below 55°C (53 or 54°C). RESULTS In the first study, the rats received TUS after dMCAO For the ATD calculation over multiple sonications, chemical-shift arti- showed significantly lower NSS (n=10, 5.5±2.5) than the Control facts in MR thermometry were corrected retrospectively in Matlab if group (n=10, 10.5±1.4) (p<0.01). In addition, the stroke rats received the misalignment between successive sonications exceeded 1 mm TUS had significantly reduced cortical infarct volume than the control by manually shifting the centre of the heating area along the group (13.78% ± 8.18%, n=16, versus 43.39%±2.33%, n=16, p<0.01). frequency-encoding direction. The ATD was then integrated in the What’s more, Immunohistochemical staining indicated reduction of axial plane using the standard thermal dose model. The spatial di- neutrophils in the affected area, and laser speckle imaging showed mensions of the ATD at 17, 40 and 240 CEM were measured and significant increase of a cerebral blood flow after TUS, which consist- correlated to the lesion sizes measured in T1- (3D FSPGR, TR 8.3 ms, ently supported the neuroprotection of pTUS in ischemic brain injury. TE 3.3 ms, slice thickness 1.2 mm) and T2- (FRFSE, TR 5200 ms, TE In the second study, the ischemic stroke rats received TUS precondi- 100 ms, slice thickness 3 mm) weighted images acquired on the first tioning had smaller ischemic areas during stroke induction, and 24 day follow-up. Logarithmic regression was applied on the correlation and 48 hours after the stroke, and smaller infarct volume (1.770 ± coefficients to find the best estimates for the thermal dose 0.169%) than the controls (3.215 ± 0.401%) (p<0.01). Moreover, the thresholds. stroke rats with TUS preconditioning experienced lower volume of RESULTS Thermal lesions were successfully produced in all 5 pa- brain edema than the control group. tients. The lesion size in the axial plane (AP and LR) as measured on CONCLUSIONS In the first study, both behavior and histological re- T2w images (5.0±1.9 mm) was 16% larger than that found on the sults suggested that TUS on ischemic core immediately after ische- corresponding T1w scans (4.3±2.0mm), the latter of which primarily mic stroke could be neuroprotective. In the second study, the results revealed vascular damage. As a result, logarithmic regression showed demonstrated that TUS preconditioning before photothrombosis that the best correlation of the ATD to the lesion size on T2w im- could provide neuroprotection by increasing the brain’s tolerance to aging was close to 100 CEM (linear regression slope=0.98, R = subsequently induced focal ischemic injury. 0.62), whereas the best correlation of the ATD to the lesion size on T1w imaging was close to 200 CEM (linear regression slope=0.94, R = 0.84). 17 and 40 CEM significantly overestimated the lesion size Posters in T1w images by 50% and 30%, respectively. These results are in good agreement with our previous study, which included 36 P1 treatments with peak temperatures mostly above 55°C. Therapeutic effects of chemo-agent delivery across the CONCLUSIONS An ATD of 100 CEM showed a good correlation to blood-brain barrier using mrgfus: in vivo study using the lesion size measured on T2w images acquired one day post treat- drug-resistant breast cancer brain metastasis model ment. This threshold appears to be valid both for lesions produced Eun-Joo Park, Yuri Cheon, Yun Deok Ahn, Jae Young Lee by sonications above 55°C, and by repeated sonications below 55°C. Radiology, Seoul National University Hospital, Seoul, Korea These results provide important guidance for predicting the lesion Correspondence: Eun-Joo Park size induced during focused ultrasound-based treatment of essential Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P1 tremor. (No Image Selected). OBJECTIVES Several studies have shown the promising results of drug delivery across the BBB using MRgFUS in combination with O86 microbubbles. However, more studies are required to evaluate this Neuroprotection of low-intensity transcranial ultrasound treatment technique prior to apply in clinic. This study was designed stimulation in ischemic stroke rats to evaluate the therapeutic effects of chemo-agent for brain J. Sun, H. Li, S. Tong, metastasis cancer by delivering drug across the BBB using MRgFUS School of Biomedical Engineering, Shanghai Jiao Tong University, andmicrobubbles. Shanghai, China METHODS In vivo studies were performed in two steps. For the first Correspondence: J. Sun step, multiple BBB-openings were performed at four different FUS Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):O86 condition along with microbubbles (SonoVue) to find safe and effect- ive treatment condition. The treatment was performed once a week OBJECTIVES Transcranial Ultrasound Stimulation (TUS) is a non- for six weeks. In the second step, breast cancer cells (BT-474) were invasive neuromodulation technique which modulates the neural ac- pre-treated with chemo-agent prior to the inoculation in the rat brain tivity by delivering specific pulsed ultrasonic waves through intact for the brain metastasis model. Animals were treated in three groups: scalp and skull to the targeted brain regions. In recent years, studies control, chemo-agent treatment only, and chemo-agent treatment have demonstrated that low-intensity TUS is a safe neuromodulation with BBB-opening. FUS condition and injection volume of microbub- technique with high focality and deep penetration depth, and has bles for BBB-opening were obtained from the first step experiment. therapeutic effects for brain diseases like epilepsy and stroke in ani- Animals were treated on a weekly basis for six weeks and post-treat- mal models. With these advantages, we have applied TUS in precon- ment tumor growth monitoring was followed for 12 weeks. ditioning and neuroprotection of ischemic rats, so as to verify the RESULTS As restuls of this study, histological evaluation of each BBB- neuroprotection effects of TUS in stroke rats. opening FUS condition in combination with SonoVue, therapeutic ef- METHODS In the first study, we applied a low-intensity TUS to the is- fects of chemo-agent treatment with BBB-opening, and the survival chemic cortex of SD rats immediately after a distal middle cerebral rate of each treatment will be presented. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 58 of 122 CONCLUSIONS This study evaluated the FUS condition for multiple P3 BBB-opening in combination with microbubbles, SonoVue. With the Focused ultrasound-enabled delivery of radiolabeled nanoclusters treatment conditions, therapeutic effects of chemo-agent delivery to the pons with concurrent pet imaging 1 4 4 4 across the BBB for brain metastasis cancer model by using breast Dezhuang Ye , Sultan Deborah , Hannah Luehmann , Yuanchun Tai , 2 4 3, 5 cancer cells pre-treated chemo-agent. Josh Rubin , Yongjian Liu , Hong Chen Mechanical Engineering and Material Science, Washington University in St. Louis, Saint Louis, Missouri, USA; Pediatrics Hematology/Oncology, P2 WashingtonUniversity in St. Louis, Saint Louis, Missouri, USA; Biomedical Optimizing passive cavitation mapping by refined minimum Engineering, Washington University in St. Louis, Saint Louis, Missouri, variance-based beamforming method: performance evaluationsin USA; Radiology - Rad Sciences, Washington University in St. Louis, Saint macaque models for blood-brain barrier disruption Louis, Missouri, USA; Radiation Oncology, Washington University in St. 1,2 1 2 1 2 Tao Sun , Calum Crake , Brian Tracey , Costas Arvanitis , Eric Miller , Louis, SaintLouis, Missouri, USA Nathan McDannold Correspondence: Dezhuang Ye Radiology, Brigham and Women's Hospital; Harvard Medical School, Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P3 Boston, Massachusetts, USA; Electrical and Computer Engineering, Tufts University, Medford, Massachusetts, USA OBJECTIVES Pontine glioma is the single greatest cause of brain Correspondence: Tao Sun tumor-related death in children. Pons is a major structure in the Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P2 upper part of the brainstem, which is involved in the control of func- tions such as breathing, hearing, taste, balance, and communication OBJECTIVES Microbubble-mediated focused ultrasound (FUS) therapies between different parts of the brain. The unique location of the pons harness mechanical and/or thermal effects to deliver drugs or ablate precludes surgical interventions while conventional chemotherapy tissues. Passive acoustic mapping (PAM) enables the spatio-temporal shows little improvement in survival because the blood-brain barrier monitoring of cavitation activity, which is critical for the clinical transla- (BBB) remains intact in pontine glioma. Recent efforts have focused tion of this technique. Traditional PAM is based ondelay-and-sum (DAS) on improvements in chemotherapeutic agents employed and in beamforming, a method whose quality tends to deteriorate due to is- methods of localized and targeted drug delivery. Several drug deliv- sues including multi-bubble interference, distortion in the wavefront ery strategies, such as convection-enhanced delivery and intranasal caused bythe presence of the skull, unmodeled variability of array ele- delivery, have been proposed to bypass the BBB for the treatment of ments, etc. To provide for robustness, here we consider the use of mini- pontine glioma but have met withminimal success. Focused ultra- mum variance adaptive beamforming toPAM and demonstrate sound (FUS) combined with microbubbles has been successfully used significant improvement in image quality compared to DAS. in the noninvasive and localized trans-BBB delivery of various drugs METHODS Experiments were performed in phantom with macaque and several nanoparticles for the treatment of brain cancer. 64Cu- skull and in vivo for monitoring FUS-induced blood-brain barrier disrup- alloyed gold nanoclusters (64CuAuNCs) composed of a few gold tion in a clinical MRIguided FUS system (ExAblate 4000, InSightec, Haifa, atoms and radiolabeled copper atoms are emerging theranostic Israel), which was integrated with a clinical 3T MRI unit (GE Healthcare). agents for cancer treatment. Their small sizes with PET imaging A 1024-element phased array was driven to transmit 10-ms pulsed FUS capability present unique advantages to be integrated with the FUS- at 220 kHz. RF data for PAM were acquired using a research ultrasound technique. The goal of this study was to demonstrate that FUS can imaging system (Verasonics, Redmond, WA) passively. The imaging enable the delivery of 64CuAuNCs noninvasively and locally to the probe (L382, Acuson, WA) was a 128-element linear array (fc = 3.21 pons with concurrent PET imaging capability for quantitative evalu- MHz; bandwidth: ~ 75%). Transition and receiving systems were syn- ation of the delivery outcome. The long-term perspective is to apply chronized and the first180 μs of RF-data were recorded for each burst. this novel image-guided drug delivery platform for the treatment of RESULTS Minimum variance distortionless response (MVDR) method pontine gliomas. was evaluated and further improved by adding diagonal loading and METHODS First, six wild-type mice were treated by FUS for evaluat- using subarray for covariance estimate. Results demonstrate the reso- ing the feasibility of safe and localized FUS drug delivery to the pons lution improvement and contrast enhancement using both trad- using a model drug: 40 kDafluorescently-labeled dextran with a itional and refined MVDR Beamformers compared toDAS. Axial full hydrodynamic diameter (~2.3 nm) that was close to that of the width at half maximum was tightened down to 79.5% and 38.5% of 64CuAuNCs (3-5 nm). Dextran delivery outcomes were evaluated that in DAS image for traditional and refined MVDR Beamformers, based on fluorescence images of ex vivo brain slices and safety was respectively. Moreover, refined MVDR method greatly enhances the assessed using histological staining. Second, another group of six robustness when traditional MVDR Beamforming induces more mice was used for demonstrating the feasibility of FUS-enabled artifacts due to the self-nulling effects (shown in Fig. 1). 64CuAuNCs delivery to the pons. FUS sonicated the left pons in the CONCLUSIONS It was demonstrated that the refined MVDR method presence of systemically administrated microbubbles. After FUS son- improved the image quality significantly compared to traditional DAS ication, mice were transferred to microPET/CT facility. 64CuAuNCs in method. We anticipate that the proposed methods will improve our 100 μL saline was injected into the mice via the tail vein, followed by ability to monitor and control FUS-induced cavitation-based therapies. PET scanning of themice at 1 hr, 4 hr, and 24 hr. To further validate the delivery of 64CuAuNCs, auto-radiography of ex vivo brain slices was performed, followed by inductively coupled plasma mass spec- trometry (ICP-MS) quantification of the gold concentration in the brain. At last, the biodistribution of 64CuAuNCs was evaluated using gamma counting and ICP-MS. RESULTS Fluorescence images of mouse brain slices showed localized delivery of the dextran at the FUS targeted left side of the pons. Hematoxylin and eosin stainingof the whole brain confirmed that the treatment did not cause histological level tissue damage. PET images showed targeted delivery of 64CuAuNCs to the pons and quantitative analysis of the PET images found the delivery efficiency was1.44%ID/g. Auto-radiography further validated the successeful delivery of 64CuAuNCs to the pons. ICP-MS quantification found 3-fold increase in gold concentration at the FUS-treated left side of the pons compared Fig. 1 (abstract P2). An example to show the -3dB profile. The with the contralateral nontreated right side ofthe pons. Biodistribution presented method can enhance the robustness of traditional MVDR study showed the 64CuAuNCs was cleared by liver and kidney, demon- when the main scatters were self-nulled strating the reduced systematic toxicity of this nanoparticle. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 59 of 122 CONCLUSIONS The unique location of the pons and the intact BBB in pontine glioma present a critical need for noninvasive and localized tech- niques to overcome theBBB. Although FUS technique has been under de- velopment for brain cancer drug delivery for more than one decade, this is the first study that demonstrated the successful drug delivery to the pons. Meanwhile, nanomedicine is the next-generation platform technol- ogy for cancer therapy, and FUS has been shown promising in the deliv- ery of several nanoparticles. This study demonstrated the unique integration of the FUS technique with nanoclusters for brain drug deliv- ery. The small sizes of nanoclusters presented unique advantages in trans-BBB delivery and brain parenchyma penetration. Last but not least, contrast-enhanced MRI is currently the standard technique for FUS-in- Fig. 1 (abstract P4). Typical ESD responses and the animal brain duced BBB opening quantification. Contrast-enhanced MRI monitors con- sections of a two animals with (a) or without (b) successfully induce trast agent leakage and assumes an indirect correlation between the BBB opening. The Red circles in (a) shows the EB stained regions delivery of the contrast agent and therapeutic agents. PET imaging of which represent the BBBopened region. (d) Comparison between radiolabeled nanoparticles provides a noninvasive, highly sensitive, and the peak ESD magnitudes with the BBB status (intact or opened). quantitative method for directly evaluating the trans-BBB delivery of n = 8 for eachgroup nanoparticles. P5 P4 Microbubble-facilitated focused ultrasound enhances the delivery An acoustic emission-feedback planar ultrasound system for of virus-like particles into the brain localized blood-brain barrier opening and monitoring Chia-Jun Lin, Hong-Wei Yang, Hao-Li Liu 1 1 1 2 Yu-Xian Lin , Yu-Chien Lin , Chih-Hung Tsai , Wen-Shiang Chen , Claude Department of Electrical Engineering, Chang Gung University, Taoyuan 3 1,4 Inserra , Hao-Li Liu City, Taiwan Electrical Engineering, Chang Gung University, Taoyuan City, Taiwan; Correspondence: Chia-Jung Lin Physical Medicine & Rehabilitation, National Taiwan University Hospital, Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P5 3 4 Taipei, Taiwan; Université de Lyon, Lyon, France; Neurosurgery, Chang Gung Memorial, Taoyuan City, Taiwan OBJECTIVES Microbubble-facilitated focused ultrasound (FUS) has been Correspondence: Yu-Xian Lin applied to transiently induce blood-brain barrier (BBB) opening nonin- Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P4 vasively and locally. Although substance with moderate size has been shown its feasibility to be delivered into the brain via this technique, OBJECTIVES In this study, we proposed a novel planar ultrasound ap- large molecular delivery (MW > 500 kDa) is still challenging and the paratus design that can provide PCD and real-time analysis, with the delivered effeminacy is still unknown. In this study, we aim to test the intention to perform real-time planar ultrasound BBB opening moni- CNS delivered efficacy of a novel large molecule, virus-like particles toring and control. (VLPs;MW = 2000 kDa), into the brain via microbubble-facilitated FUS METHODS A planar ultrasound probe integrating with a lateral mode BBB opening, and investigate the penetration and distribution. transducer ring was coaxially arranged, which the former one is to METHODS A total of 12 ICR mice were employed in this study. A 400 perform energy exposure and the later one is to passively receive kHz spherically focused transducer was used to deliver ultrasound the backscattered emission and characterize the subharmonic/ultra- energy (0.155 MPa; burstlength = 10 ms, PRF = 1 Hz, duration = 60 harmonic emission spectrum density (ESD) during treatment. Invivo s). The focal zone was located at left cerebral hemisphere with 2-3 tube phantom experiments were conducted to characterize the de- mm depth. Brains were harvested at 4 hours post FUS treatment, pendence of ESD change with microbubble infusion. In-vitro experi- and the frozen samples were sectioned and observed with a fluores- ments were employed to characterize the dependence of ESD cent microscope (TissueFAX Plus, Austria). Tissue slices were then change, and in-vivo experiments were employed to characterize the stained byimmunofluorescence and observed to investigate the dis- effect to induce BBB opening. Evans blue extravasations and HE tribution of VLPs in brain cells. GFP-bounded dextrans with the size staining was conducted to provide histological confirmation. ranging from 70 – 2000 kDa were employed as another molecular RESULTS This study demonstrates the capability in using planar ultra- surrogate and compare with the VLP delivery. Evans blue (EB) was sound system to open the BBB. The ESD response well corresponds used as the indicator of BBB opening. to the animal brain section, where the EB well stained on the ultra- RESULTS FUS exposure with the presence of microbubbles were shown sound exposure position to induce successful BBB opening. For ani- to be able to locally open the BBB at the target site. VLP has shown evi- mal group with BBB still intact after exposure, the peak ESD was dence to be delivered into the brain (Fig. 1). When correlated with the observed to be 0± 3 dB. On the other hand, in animal group with observation of GFP-bounded dextrans, it was observed that 70 kDa dex- successful BBB-opening, the peak ESD was observed to be signifi- trans spread in the left hemisphere, while 2000 kDadextrans aggregated cantly higher (17± 12 dB) than the BBB-intact groups (Fig. 1). The 4- at some spots. More VLPs are distributed near the hemorrhage sites in dB ESD level was found to be a valid threshold level to well discrim- the left hemisphere after treated with microbubble-facilitated FUS, com- inate the BBB-intact and BBB-opened group. With the integration of paredto the non-hemorrhage sites. VLPs are co-localized with neuron nu- the lateral-mode ring transducer, the proposed system also demon- clei (NeuN), while not obviously with gilal fibrillary acidic protein (GFAP). strates the feasibility to monitor the BBB-opened status, making the 2000 kDa dextranaggregated between the brain cells but showed no evi- real-time control of the process become feasible. dence of the entry in brain cells. CONCLUSIONS This study may provide valuable information toward CONCLUSIONS In this study, we have demonstrated that large mo- designing a planar ultrasound treatment apparatus for the purpose lecular delivery up to 2000 kDa is feasible when combining with of BBB opening and brain drug delivery. microbubble-facilitated FUS BBBopening. We observed that the VLPs Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 60 of 122 when entering into brain is able to penetrate into neuron cells, sup- the results of western blot were calculated on amounts of P-gp com- port the VLP-cell endocytosis is occurred. VLP preserve superiority in pared with β-actin. In addition, the change of P-gp in sonicated area its high endocytosis feature, making the VLP has the potential to shows percent assuming that P-gp of non-sonicated area is100%. At serve as a gene carrier to perform CNS gene delivery. 1 hour, P-gp of sonicated region was reduced by 15% and showed the lowest reduction on 1 day (76%). From 2 to 5 day, P-gp seemed to be recover gradually. It was reduced by 61% on 2 day and 43% on 3 day. On 5 days, it was completely recovered and showed no de- crease (Fig. 1B). CONCLUSIONS The results of this study provide the information needed to take into account the dynamics P-gp change over time after FUS-induced BBB disruption. Therefore, these results could sug- gest more detailed treatment protocols when FUS-induced BBB dis- ruption treatment along with P-gp substrate drug. Fig. 1 (abstract P6). MR image (A), Pglycoprotein quantification graph (B) Fig. 1 (abstract P5). (A) After microbubblefacilitated FUS treatment P7 at left hemisphere, EB entered into the brain, indicating the BBB is Focused ultrasound-induced blood-brain barrier opening enhances opened. Some hemorrhage was observed. One unit of scale is 1 GSK-3 inhibitor delivery for amyloid-beta plaque reduction mm. (B) VLP fused with mcherry fluorescent protein entered into the 1,2 2 3 4,5 PoHung Hsu , Yi-Hsiu Chung , KunJu Lin , LiangYo Yang , left hemisphere after BBB opening induced by microbubblefacilitated 2 1,6 Tzu-Chen Yen , Hao-Li Liu FUS.VLP was shown as red. Nuclei were stained by DAPI. (C) Compared 1 2 Electrical Engineering, Chang Gung University, Taoyuan City, Taiwan; Center with saline injection or VLP injection without FUS, more VLPs were forAdvancedMolecular ImagingandTranslation, ChangGungMemorial colocalized with NeuN after BBB opening, indicated its entry into the Hospital, Taoyuan City, Taiwan; Department of Nuclear Medicine, Chang neuron cells. VLP was presented as green, NeuN was as red and nuclei Gung Memorial Hospital, Taoyuan City, Taiwan; Department of Physiology, was as blue School of Medicine, College of Medicine, China Medical University, Taichung, Taiwan; Department of Biotechnology, Asia University, Taichung, Taiwan; Medical Imaging Research Center, Institute forRadiological Research, Chang P6 Gung University and Chang Gung Memorial Hospital, Taoyuan City, Taiwan The kinetics of P-glycoprotein after blood-brain barrier Correspondence: PoHung Hsu disruption in rat brain by MRI-guided focused Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P7 ultrasound Mun Han, Danbi Kim, Sanghyun Ahn, Juyoung Park OBJECTIVES Alzheimer's disease (AD) is a chronic neurodegenerative Daegu-Gyeongbuk Medical Innovation Foundation, Daegu, Korea disease that is the leading cause of age-related dementia. Currently, Correspondence: Mun Han therapeutic agent delivery tothe CNS is a valued approach for AD Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P6 therapy. Unfortunately, CNS penetration of therapeutic agents is greatly hampered by the blood-brain barrier (BBB). Focused ultra- OBJECTIVES Multi-drug resistant proteins in brain endothelial cells sound (FUS) exposure has been demonstrated to temporally open pump drugs in brain tissues out into the blood and could diminish the BBB, thus promoting therapeutic agent delivery to the CNS for drug retention and drug efficacy. In previous our study, a representa- AD therapy. In this study, we aimed to evaluate whether the use of tive multi-drug resistant protein, P-glycoprotein (P-gp), was signifi- FUS-induced BBB opening to enhance GSK-3 inhibitor delivery can cantly reduced at 24 hrs after blood brain barrier (BBB) disruption by promote Aβ plaque clearance and synthetic regulation in a trans- focused ultrasound (FUS). In this study, we investigated the kinetics genic small animal model (Fig. 1a). of the P-gp after FUS-induced BBB disruption. METHODS FUS-induced BBB opening on APP/PS1 transgenic mice METHODS The rat brain in thalamus was sonicated (0.5~0.55 MPa) was performed unilaterally, with the contralateral hemisphere serving transracially using a 1 MHz FUS transducer with 10 ms bursts of ultra- as a reference. AV-45PET/CT imaging was attempted to detect plaque sound wave at 1Hz pulse repetition frequency for 120s, combined distribution and concentration in vivo, and autoradiography (ARG) with intravenous injection of a microbubble ultrasound contrast was conducted ex vivo to quantitatively confirm theAβ plaque reduc- agent (Definity; 400 μl/kg). T1 and T2 weighted MRI scan were used tion. Immunohistochemistry staining was also performed to confirm to guide FUS sonication into the targeted brain and confirm the BBB the GSK-3 expression level. disruption. Then, the sonicated rat brains were extracted at different RESULTS Results from AV-45 PET/CT imaging showed positive Aβ time points (1 hr, 1, 2, 3, 5day) after BBB disruption. In order to meas- plaque regulation in vivo,and ex vivo autoradiography (ARG) further ure the P-gp expression, choroid plexuses were removed from all showed significant radiolabelled tracer detectability (up to 26.72% Aβ ventricles and then western blot analysis was performed using plaque reduction). Immunohistochemistry revealed that the GSK-3 in- antibody against P-gp (1:100) and β-actin (1:2000). P-gp expression hibitor effectively blocked plaque synthesis up to60.6% in the GSK-3 sonicated area and non-sonicated area were compared. immunoreactive area, confirming the proposed therapeutic pathway. RESULTS The BBB disruption was confirmed in the thalamus region CONCLUSIONS We demonstrated that FUS-induced BBB opening to through T1 weighted images with MR contrast agent (Fig. 1A). All enhance GSK-3 inhibitor delivery can efficiently reduce amyloid-beta Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 61 of 122 plaques in transgenicmouse brains. The results showed that the use Bracco, 500 ul/kg) using a 650 kHz transducer (focal spot 3.5 x 22 mm). of AV-45 PET imaging can serve as a diagnostic tool for in vivo quan- Subsequently the FUS set-up was replaced with a set-up for 13C-im- titation of plaque clearance, while ex vivo autoradiography (ARG) can aging, containing a dedicated 13C surface coil inside a small 1H volume provide radio-labelled tracer detectability. This study improves our coil for anatomical reference imaging. An interesting metabolite to understanding of how ultrasound can be used to enhance AD thera- study is pyruvate, which is the product of the glycolysis of glucose and peutic molecule delivery, and promote advances in discovering new is further metabolized in the Krebs’s cycle or converted tolactate. There- therapeutic strategies for neurodegenerative disease. fore, [1-13C] pyruvate was polarized in an in-house-built polarizer as de- scribed previously [Breukels, 2015]. The polarized substrate was rapidly dissolvedin a 1xPBS buffer solution to a final concentration of 80mM. 300μl HP pyruvate was injected within 15s after dissolution. To enable dynamic imaging of the conversion of pyruvate to lactate, a slice select- ive gradient echo sequence was developed that makes it possible to simultaneously image pyruvate and lactate based on their chemical shift difference [Steinseifer 2013]. Every 1.6s an image is acquired that consists of a pyruvate and lactate image [Example, Fig. 1]. A phantom containing [1-13C] pyruvate and [1-13C] lactate was placed next to the animal to obtain a 13C reference image for image registration. From these images, concentration curves of pyruvate and lactate can be ob- tained that by plotting the total signal intensity over time of a region of interest. These concentration curves show the conversion of pyruvate to lactate (Example, Fig. 2). Since gadolinium-based contrast agents Fig. 1 (abstract P7). (A) The conceptual schematics of this study. (B) cannot cross the intact BBB, Magnevist (0.25 ml/kg) was injected after ThioflavinS staining and quantification of amyloidbeta plaque sizes 13C-imaging to verify opening of the BBB using pre- andpost-contrast in the hippocampus area. There was a7.83% (FUS alone), 14.68% (AR T1-weighted MR imaging. alone) and 26.72% (AR+FUS) decrease in comparison to the control RESULTS Pre- and post-contrast T1-weighted images in Fig. 3 confirm group. AR = ARA014418; FUS = focused ultrasound (Fig. 1b) a successful opening of the BBB, as we observed a hyperintense re- gion were the brain was sonicated. We successfully obtained pyru- vate signal in the 13C images, but the intensity of the lactate signals was too low to observe. The pyruvate image did not show any evi- P8 dent signal intensities that can be related to BBB-opening. Novel approach to study metabolic changes after FUS-mediated CONCLUSIONS In this study we successfully showed the feasibility of blood-brain barrier disruption combining BBB disruption using FUS with dynamic imaging of hyper- Thiele Kobus, Tom Peeters, Andor Veltien, Arend Heerschap, polarized 13C-labeled compounds. The BBB was disrupted prior to, and Tom Scheenen remained open during our hyperpolarized 13C-imaging experiment, as Radboud university medical center, Nijmegen, Netherlands confirmed by contrast-enhanced MRI. However,SNR of the hyperpolar- Correspondence: Thiele Kobus ized signals is currently too low to draw conclusions upon enhanced Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P8 uptake or release of hyperpolarized metabolites in the targeted brain region.Therefore, improvement of polarization levels and further OBJECTIVES Focused ultrasound (FUS) in combination with micro- optimization of the imaging parameters is necessary.Despite challenges bubbles can be used to temporarily and locally disrupt the blood- that have to be overcome, this approach enables delivery of hyperpo- brain barrier (BBB) [Hynynen, 2001]. However, little is known about larized agents to the brain and will allow us to study metabolic alter- the exact mechanism of this technique and its effects on the brain. ations due todisruption of the BBB in vivo in the near future. To get more insight, we would like to study metabolic changes after disruption of the BBB. Many metabolites contain carbon-atoms of which approximately 1% is the isotope 13C that can be detected by magnetic resonance imaging (MRI). Due to the low abundance and low gyromagnetic ratio, the sensitivity of 13C MRI is low. Recently, a method has become available to increase the sensitivity of13C-la- beled substrates more than 10.000 fold by dynamic nuclear hyperpo- larization (DNP) and create a solution that can be injected [Ardenkjaer-Larsen, 2003]. After injection of the sample, the conver- sion of the substrate into other metabolites can be observed with 13C MRI enabling the study of fast dynamic metabolic processes in vivo. In this proof of concept, we demonstrate the feasibility of combining FUS-mediated disruption of the BBB with hyperpolarized 13C MRI using DNP. In the near future, this approach might reveal the influence of the BBB on the uptake of hyperpolarized agents or alterations in metabolism due to FUS-mediated BBB opening. METHODS The experimental protocol was approved by the National Animal Research Authority. In two mice, the BBB was disrupted using FUS and followed by hyperpolarized 13C MRI. All experiments took place on a 7T animal system (Clinscan, Bruker). For BBB disruption, animals were anesthetized and positioned in a dedicated MR-guided FUS set-up (IGT, France). MR reference images were acquired for Fig. 1 (abstract P8). Example of a hyperpolarized pyruvate and ultrasound targeting in the left frontal lobe of the brain. The sonic- lactate image mapped to the T2-weighted reference image (no ation (duration=120s, 10ms bursts, burst frequency=1Hz) was per- BBB disruption) formed immediately after the injection of microbubbles (Sonovue, Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 62 of 122 presents control kernel of the system, multiple channels power amplifier, high speed power sensing circuits, low pass filters, and a module of matching circuits. The FPGA-based digital board has three functions. Firstly, it is used as a communication unit with PC through an USB3.0 interface (CYUSB3014, Cypress, USA). Sec- ondly, it can monitor the excitation power by incorporating high speed power sensing module. Lastly, the most important function for the board is to generate multiple channel source waveform and each channel waveform can be controlled individually. The phase information of the waveform is flexible to change so that the system can implement different beamforming algorithms. The power amplifier board is used to amplify the low voltage wave- form to drive the transducer. The low pass filter is connected with the power amplifier board to shape the high-voltage wave- form to match with transducer. A matching circuit is employed to adjust the impedance characteristics to achieve maximal power transfer. And a dynamic phased control beamforming al- gorithm is developed to achieve flexible beam focusing in the acoustic field. RESULTS The prototype of the fabricated ultrasound system is shown Fig. 2 (abstract P8). Example of a dynamic time evaluation of in Fig. 1b. Each module is achieved by separate printed circuit boards hyperpolarized pyruvate and lactate signals (no BBB disruption). The (PCBs). The phase difference error for all channels is less than 7 ns total signal intensity of a region of interest isplotted over time. Data when the center frequency of transducer is set to 800 KHz. It means were not corrected for T1decay and signal loss due to RF pulsing that the phase error of the proposed ultrasound systemis about 2 de- grees, which is able to support a good beamforming performance. Dynamic focusing can be achieved in a circle with a radius of 10 mm, and the focal depth can be adjusted in the range of 4-10 cm. The system can work in a pulsed mode or a continuous mode for 50- 300 ms. The system is able to hit a water column which theheight is about 5 cm. The beamforming quality of the system is also evaluated by an acoustic hydrophone. CONCLUSIONS In this study, we present a prototype design of a low cost ultrasound array system specifically for neuromodulation. Test re- sults demonstrate that the proposed system has a great potential to Fig. 3 (abstract P8). Pre (A) and postcontrast (B) images after gadolinium improve the level of neuromodulation. Future work such as improve- administration. The subtraction image (C) shows a hyperintense region ment of beamforming performance in different acoustic mediums and where the BBB was successfully disrupted. The corresponding 13C image (D) validation of primate animal studies in vivo will be carried out. of hyperpolarized pyruvate after BBB disruption P9 A low-cost phased array system for ultrasound neuromodulation Wu Sun, Juan Zhou, WenBin Yan, JiaXing Yang, Weibao Qiu, Hairong Zheng Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences, Shenzhen, China Correspondence: Wu Sun Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P9 OBJECTIVES Ultrasound has been shown to non-invasively stimulate Fig. 1 (abstract P9). (a) The schematic of the ultrasound electronic and inhibit neuronal activities under various conditions like stimulat- system. (b) the prototype of the proposed ultrasound system ing auditory nerve responses and suppressing sensory-evoked poten- tials in the primary visual cortex. Currently most of the published works for ultrasound neuromodulation are based on single element focused transducer. In contrast with single element transducer, P10 phased array transducer is more flexible to change the target loca- Test study for the electro-acoustic conversion efficiency of focused tion, and potentially to simulate multiple positions simultaneously. ultrasonic transducer based on virtual instrument 1 2, 1 1, 3 1 High intensity focused ultrasound (HIFU) array system can be applied Yang Liu , Jianwen Tan , Deping Zeng , Zhiming Zhong to neuromodulation by decreased the ultrasound energy, however, Biomedical Engineering, Chingqing Medical University, Chongqing, such a system is usually quite expensive and bulky. Thus a low cost China; Chongqing Communication Institute., Chongqing, China; ultrasound array system specifically for ultrasound neuromodulation National Engineering ResearchCenter of Ultrasound Medicine, is needed. This paper presents an ultrasound phased array system Chongqing, China that incorporates a 256-element hemispheric transducer for neuro- Correspondence: Yang Liu modulation. Test results show that the system has potential to Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P10 achieve multiple position neuromodulation. METHODS Our system includes two parts: 1) ultrasound electronic OBJECTIVES The electro-acoustic conversion characteristic of ultrasonic system; 2) phased array transducer (256-element array transducer, transducer is required to be tested when developing focused ultrasound Imasonic, France). The schematic of the ultrasound electronic sys- treatment equipment, with the actual electro-acoustic characteristics of tem is shown in Fig. 1a. The system consists of five major circuit transducer to determine working frequency and driver parameters of fo- modules: an field programmable gate array (FPGA) (Cyclone cused ultrasound therapy system. Therefore, when the HIFU therapy sys- V5CGXFC7D7F31C8, Altera, USA) based digital board which tem developed, there is a massive workload in the measurement of Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 63 of 122 electro-acoustic conversion characteristics of focus ultrasonic transducer. It has put forward higher request to the efficiency and convenience of the electro-acoustic conversion characteristics test system. METHODS An automatic test system for the electro-acoustic conversion characteristic of the transducer based on virtual instrument has been de- veloped in this letter. The input electric power of the transducer is mea- sured by electric power meter based on directional coupler, radiation force balance is used to measure the output soundpower of the trans- ducer. By use of the graphic development environment LabVIEW, an upper computer automatic test software was developed based on virtual instrument technology, which can operate with the real-time data acqui- sition and data processing, analysis and calculation of the data. RESULTS Based on this test system, the electro-acoustic conversion efficiency of the focused ultrasound transducer at different frequencies and the driving power also tested in this paper, the bandwidth of the transducer was ana- lyzed from the point of view that the measured electro-acoustic conversion efficiency of transducer, which discussed electro-acoustic conversion efficiency of transducer with the tracking problem of different driving power and oper- ating frequency based on electrical reflectioncoefficient (Figs. 1, 2, 3). CONCLUSIONS With the analysis of test results shows that the proposed Fig. 3 (abstract P10). The reflectance of transducer under different real-time monitoring tool based on electrical reflection coefficient which driving electric power measured by directional coupler can also applied to tracking the fre- quency of the ultrasonic transducer, which can improve the electrical power transmission efficiency and the output acoustic power of ultra- P11 sonic transducer, and there is important significance on the focused ultra- Multifunctional pulse generator unit for high-intensity focused sound transducer to ensure safety, stability and efficient operation. ultrasound system 2,1 2 2 Satoshi Tamano , Shin Yoshizawa , Shin-ichiro Umemura 1 2 Hitachi, Ltd., Kokubunji, Japan; Tohoku Univ., Sendai, Japan Correspondence: Satoshi Tamano Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P11 OBJECTIVES High-intensity focused ultrasound (HIFU) therapeutic systems have made it possible to treat diseases with fine spatial reso- lution. One critical technical challenge of HIFU is to generate high- voltage, large-current, multi-channel pulsed signals to effectively ex- cite the low-impedance HIFU transducers. This paper presents the development of multifunctional pulse generator unit for HIFU system. The pulse generator can produce pseudo sinusoidal wave, triggered- HIFU mode pulse and second-harmonics superimposed shock wave lithotripsy pulses for HIFU transmission. We report the development of a novel multifunctional, reconfigurable pulse generator using FPGA and high-voltage MOSFETs. METHODS The pulse generator is interfaced with a PC and receives commands, waveform data, focusing delay data through a high-speed USB 2.0 micro controller. The USB microcontroller transmits communi- cation information with the PC to the control FPGA. The control FPGA Fig. 1 (abstract P10). The electro-acoustic conversion efficiency and stores instruction information from the PC internalregisters, and de- reflectance of focused ultrasound transducer at different frequencies livers transmission waveform information and transmission focus data to the transmission FPGA. For data communication between the con- trol FPGA andthe transmission FPGAs, information is transferred using a dedicated local bus. In our system, 32 channels transmission can be performed per transmission board. Sincewe use HIFU transducer con- sisting of 128 elements, we used four transmission boards. The control board performs generation of actual ultrasonic wave sequences and ar- bitration of the transmission board control bus as well as the functions described above. The transmission board includes a transmission FPGA for generating a timing signal for ultrasonic transmission based on a command transmitted from the control FPGA, RAMs that stores trans- mission focus data, N-channel and P-channelMOSFETs that actually generates 400 Vpp ultrasonic transmission signal, MOSFET drivers for driving MOSFET. In this study, the operation of the transmission circuit- was evaluated using our prototype dual transmission frequencies trans- ducer. This prototype transducer consists of seven elements that can transmit one and two MHz ultrasonic waves. With these two transmis- sion frequencies, since the impedance of the transducer is around 150 Ω, it is necessary to apply a transmission circuit withlarge driving capability. RESULTS In pseudo sinusoidal transmission waveform mode, the test results show that the power consumption is reduced by 14.8 % and Fig. 2 (abstract P10). The electro-acoustic conversion efficiency and MOSFET maximum temperature rise is reduced by 11.5 °C by using reflectance of transducer under different driving electric power the newly proposed circuit than the our previous class D circuit. Also, Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 64 of 122 it can be seen that the third and fifth harmonic components can be P13 suppressed by 23.9, 30.5 dB, respectively. Therefore, the transmission Axial controllable multiple traps of acoustic vortices generated by waveform becomes closer to the sinusoidal waveform. As a result of directional sources 1 1 1 2 2 the above, we quantitatively evaluated the power saving effect by in- Yuzhi Li , Gepu Guo , Qingyu Ma , Juan Tu , Dong Zhang 1 2 creasing the number of power supply steps and the effect of reducing Nanjing Normal University, Nanjing, China; Nanjing University, Nanjing, the device temperature rise. In the triggered-HIFU mode, high power China (>300 Vpp) and a short time (several microseconds) ultrasonic radiation Correspondence: Yuzhi Li called trigger-burst, and a medium output (< 100 Vpp) andlong dur- Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P13 ation (several seconds) ultrasonic radiation called heating-waves are used in combination. In the transmission circuit, heating-waves has a OBJECTIVES Characterized by the pressure circle and phase spiral, larger amount of heat generation than trigger-burst, so there was a big acoustic vortex (AV) can be used to manipulate objects with its or- problem how to realize a circuit with reduced device power con- bital angular momentum androtation torque. Compared with light, sumption. The proposed circuit can 18.9% lower power consumption acoustic wave can go into media non-destructively with deeper than conventional class D circuit, and 33.6 °C suppress MOSFET penetration depth, which makes it possible to manipulate particles temperature rise. In peak negative enhanced second-harmonics super- inside object using AV, exhibiting a prosperous future in biomedical imposed shock wave lithotripsy mode, by using the proposed circuit, it engineering. In previous studies, AV was often investigated under was possible to realize a concavity of the positive voltage side output, the framework of point source radiation with acoustic diffraction. and the second harmonic ratio approached thetheoretical value 2.0 dB However, the point source based model is not practical for direc- from the class D circuit. tional sources with big ka values, which are influenced by beam- CONCLUSIONS We modified the HIFU ultrasonic transmission circuit patterns with obvious side lobes. Consequently, more attentions previously reported at ISTU 2016 and created a HIFU transmission should be focused on the distributions of AV generated by direc- unit with 128 channelsintegrated. In particular, by increasing the tional sources. transmission voltage level from four levels to seven levels, MOSFET METHODS The phase coded approach is employed to generate con- device heat generation could be greatly reduced. At the same time, trollable AV using directional sources. Based on the radiation theory we achieved a significant reduction in power consumption. As a re- of planar piston source andcoded initial phase, the physical mechan- sult, the risk of MOSFET damage is reduced, and HIFU transmission ism of AV is theoretically investigated with explicit formulae. The could be executedstably and safety. In our newly proposed circuit, it principles of main-AV (M-AV) and vice-AVs (V-AVs) generated by the was demonstrated that it is effective not only for transmission of main lobes and the side lobs of the sources are analyzed. The num- pseudo sinusoidal wave but also for triggered-HIFUmode and sec- ber and locations of the formed M-AV, V-AVs and the corresponding ond-harmonics superimposed shock wave lithotripsy. The applicabil- vortex valleys (VVs) are calculated for different ka values. And the ity of our proposed circuit has expanded. generations of axially controllable multiple traps are discussed based on Gorkov’s theory. The proposed theory isalso verified by numerical studies and experimental measurements for different ka values at P12 the frequency of 1 MHz. The improvement effect of magnetic microbubbles on HIFU- RESULTS It is proved that obvious M-AV, V-AV and VVs can be gener- induced hyperthermia effect ated for higher ka with the least value of 3.83. Corresponding to the Dongxin Yang, Yanye Yang, Guangyao Xu, Xiasheng Guo, Juan Tu, Dong 8-source experimental systemfor ka=29.32, the measured axial pro- Zhang files and the cross-sectional distributions at different heights show Nanjing University, Nanjing, China good agreements with the simulated ones, and obvious phase spirals Correspondence: Dongxin Yang of M-AV and V-AVs are also demonstrated. Several VVs with almost Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P12 pressure zero are also observed on the center axis to form multiple traps, which can beaxially controlled for ka adjustment. OBJECTIVES Encapsulated microbubbles (MBs) have been widely CONCLUSIONS It is demonstrated that, for bigger ka, besides M-AV used to enhance high intensity focused ultrasound (HIFU) -induced generated by the main lobes of the sources, cone-shaped V-AVs pro- hyperthermia by making use of its ability of increasing acoustic en- duced by the side lobes are closer to the source plane at relatively ergy absorption and lowering the cavitation threshold. To balance lower pressure. Corresponding to the radiation angles of press valleys the needs of treatment efficiency and safety, there is increasing de- between the main lobe and the side lobes of sources, VVs with al- mand of more efficient encapsulated MB agents that can quickly most pressure zero can be generated on the central axis to form ax- achieve sufficient hyperthermia effect while minimizing the damage ially controllable multiple traps with the locations controlled by ka to surrounding tissues. adjustment. The results provide the feasibility of deep-level control METHODS In the present work, superparamagnetic iron oxide nano- of AV inside object and suggest the application potential of multiple particles (SPIO) were coupled to perfluorocarbon-filled, albumin-en- traps for particle manipulation in biomedical engineering. capsulated microbubbles (referred as SPIO-albumin MBs) to enable a stronger enhancement of the HIFU-induced hyperthermia effect than regular albumin-encapsulated ones. The thermal enhancement cap- P14 acity of SPIO-albumin MBs and albumin-encapsulated MBs was inves- Study on the acute injury effect on candida albicans by low- tigated based on both experimental measurements and numerical frequency and low-intensity ultrasound simulations ofthe temperature change at HIFU focus. Yang Xiang RESULTS The results show that, comparing with the use of albumin-en- Chongqing Medical University, Chongqing, China capsulated MBs, the adoption of SPIO-albumin MBs will bring about Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P14 quicker temperatureelevation rate and higher peak temperature. CONCLUSIONS An improvement can be made to the enhancement OBJECTIVES To investigate the effects of Low-frequency and Low-inten- of the HIFU-induced hyperthermia by using SPIO-albumin MBs rather sity Ultrasound (LFLIU) on the acute injury of candida albicans, and to in- than albumin-encapsulated ones, which is because the thermal prop- vestigate the effect of LFLIU on the permeability of the cell wall. erties of the two kinds of microbubbles are different. With these ad- METHODS Concentration is 1.5 x 107 cfu/ml of Candida albicans bacteria vantages, these SPIO-albumin MBs can be introduced tospecific liquid, with 5 ml bacteria to single flageolet culture plate. With the fre- locations of interest to intensify the thermal effect of HIFU much quency for 42 kHz, probe diameter is 5 cm circular planar ultrasonic treat- more efficiently, which might enable more ideal non-invasive con- ment head, ultrasonic intensity was selected 0.13 W/cm2, 0.35 W/cm2 trollable hyperthermia treatment strategies and applications. irradiation six well culture plate beads bacterium solution for 5 min, After Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 65 of 122 48 h count ultrasound irradiation on petri dish culture survival of colonies of transducer on MAT-MI and provide the fundamentals for transducer in a petri dish. Transmission electron microscope and scanning electron selection in further study to improve the accuracy of electrical imped- microscope were used to observe the external shape and internal struc- ance reconstruction. ture of the bacteria. RESULTS Different doses of ultrasound irradiation 5 min, 48 h after culture dish colony average count shows, Control group colony P16 count to 21 cfu, ultrasonic intensity0.13 W/cm2 group of colony A method for evaluating the relationship between the inertial count for 20 cfu, ultrasonic intensity of 0.35 W/cm2 group of colony cavitation duration and the acoustic parameters 1 1 2 1 count for 14 cfu.LFLIU candida albicans, scanning electron microsco- Mouwen Cheng , Yutong Lin , Alfred C. Yu , Peng Qin pe(SEM) visible thalli were swollen shape becomes large, Under the Department of Instrument Science and Engineering, Shanghai Jiao scanning electron microscope (SEM) visible thalli was swelling, the Tong University, Shanghai, China; Department of Electrical and shape becomes large, the extracellular fluid into the cells increased Computer Engineering, TheUniversity of waterloo, Waterloo, Alberta, obviously; Transmission electron microscope (TEM) see nuclear mat- Canada ter at the edge of the aggregation, cell membrane damage, forma- Correspondence: Mouwen Cheng tion of vacuoles, large amounts of water into cells. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P16 CONCLUSIONS With the increase of ultrasonic irradiation intensity, signifi- cantly increased the mortality rate of Candida albicans; Candida albicans OBJECTIVE Inertial cavitation, triggered by the ultrasound and micro- in LFIU can promote the increase of the permeability of the cell wall of. bubbles, is considered to be the main mechanism for sonoporation- mediated macromolecule delivery. Inertial cavitation dose in most studies has been employed to evaluate the delivery efficiency and treatment efficiency. However, the temporal characteristic of the iner- P15 tial cavitation are also closely related to the bioeffects accompanied Transducer directivity influence on artifacts reduction for by sonoporation. This study aims to propose a method for evaluating magnetoacoustic tomography with magnetic induction the temporal duration of inertialcavitation and determine its relation 1 1 2 2 Gepu Guo , Qingyu Ma , Juan Tu , Dong Zhang with the acoustic parameters. Nanjing Normal University, School of Physics and Technology, Nanjing, METHODS An agarose-gel tissue-mimicking phantom was fabricated to Jiangsu, China; Nanjing University, Nanjing, China hold 1% SonoVue microbubbles solution. 1-MHz plane transmission Correspondence: Gepu Guo transducer and another 7.5-MHzfocused transducer were employed for Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P15 triggering the cavitation of microbubbles and passively detecting acous- tic signal, respectively. Then the signal was amplified and recorded by a OBJECTIVES As a novel noninvasive modality of detecting electrical high-speed digitizer. After the frequency domains characteristics of the conductivity variation for tissues, magnetoacoustic tomography with signals were analyzed, the distribution of the broadband energies during magnetic induction (MATMI) has been demonstrated to have the cap- the exposure time was obtained (Fig. 1). The full width at the half max- ability of distinguishing the early pathological changes of biological tis- imum in the time trace of broadband energy was proposed to deter- sues inside the object. In previous studies, the transducer was usually mine the temporal duration of the inertial cavitation behavior. simplified as an ideal or omnidirectional receiver without the consider- RESULTS This study determine: 1) Peak rarefactional pressure (PRP) was ation of its directivity and scanning radius. However, the properties of negatively correlated with inertial cavitation duration. Inertial cavitation transducer play a vital role in signal acquisition and image reconstruc- duration rapidly decreased from about 29.31 ms to 1.50 ms for the in- tion. In order to optimize image reconstruction and eliminate the creasing PRP from 0.5 MPa to 0.7 MPa. But for the PRP above 0.7 MPa, in- image artifacts for MAT-MI, the influence of transducer was investigated ertial cavitation duration approximately approaches to 1.26 ms, suggested theoretically for a two-layer eccentric spherical tissue model based on inertial cavitation duration tended to be saturated. 2) Inertial cavitation the principles of acoustic dipole radiation and transducer reception. duration gradually increased from 0.60 ms to 1.23 ms with the increasing METHODS Based on the principles of magnetic excitation, acoustic vi- pulse duration (PD) from 10 μs to 400 μs, and was positively correlated bration, acoustic dipole radiation and transducer reception, numerical with the PD. 3) Pulse repetition frequency (PRF) exhibited relatively weak- simulations are performed for a two-layer eccentric spherical phantom endependent than PD on the inertial cavitation duration (Figs. 2, 3). model. The factors that affect transducer directivity are analyzed, and CONCLUSIONS The proposed inertial cavitation duration could be the distributions of acoustic pressure and waveform are simulated used to evaluate the kinetics of the inertial cavitation triggered by using the transducers with omni-directivity, strong-directivity and uni- pulsed ultrasound and microbubbles. The relationship between directivity. Then the MAT-MI images reconstruct with the collected acoustic parameters (PRP, PD, PRF) and inertial cavitation duration of acoustic waveforms are achieved and compared to the corresponding SonoVue microbubbles were determined. These finding suggested model to analyze the affect of artifacts reduction. inertial cavitation duration, in combination with the previous inertial RESULTS It is demonstrated that the image artifacts of MAT-MI is obvi- cavitation dose, would be the important factors for evaluating cavita- ous for waveform collection using omni-directional transducer. Accord- tion-mediated therapy. ing with the increase of transducer directivity, the detection angle of the receiver decreases with an increased sensitivity. Especially for a uni-direc- tional transducer, the collected pressure reflectsthe strength of acoustic vibration along the normal direction of the receiver, which can be used to reconstruct the conductivity contrast image without artifacts. In prac- tical applications, large-radius transducer with strong directivity can be applied as an approximate uni-directional receiver to improve image quality with little artifacts aspossible. In addition, to realize narrow detec- tion scope with a fixed transducer, the scanning radius should also be optimized to achieve acceptable signal to noise ratioand peak pressure ratio. The favorable results confirm the influence of transducer on MAT- MI and provide the fundamentals for transducer selection in further study to improve the accuracy of electrical impedance reconstruction. CONCLUSIONS It is concluded that large-radius transducer with strong directivity can be applied as an approximate uni-directional receiver to improve image quality with little artifacts as possible. In addition, to realize narrow detection scope with a fixed transducer, the scanning ra- dius should also be optimized to achieve acceptable signal to noise ra- Fig. 1 (abstract P16). See text for description tio and peak pressure ratio. The favorable results confirm the influence Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 66 of 122 P17 by time reversal method using simulation. At first the sound source Sub-wavelength focal region achieved by a spherical focused is set at the target point at the simulation model constructed by ultrasound transducer with open ends in resonator modes monkey’s CT data and the ultrasound emitted and propagates to the 1 1 2 1 3 3 Hua Cao , Min He , Zhou Lin , man luo , Guangrong Lei , Xiaobo Gong , array transducer through the skull. Next the received signals are re- 4 1 1 5 2 Jun Dang , Deping Zeng , Faqi Li , Junru Wu , Dong Zhang , versed and emitted from the array transducer to the target, which Zhibiao Wang focus the ultrasound on the target correctly. Simulation solves the 1 2 Chongqing medical university, Chongqing, China; Institute of basic equation of continuum mechanics by FDTD method. And it an- Acoustics, Key Laboratory of Modern Acoustics, MOE, Nanjing University, alyzes the ultrasound propagation by treating the medium as the Nanjing, China; National Engineering Research Center of Ultrasound mixture of water and bone. In order to model the complicated living Medicine, Chongqing, China; Department of Oncology, 1st Affiliated body tissues Hounsfield unit of CT images is translated to the volume Hospital of Chongqing Medical University, Chongqing, China; Physics, fraction of bone. The density and sound speed of each voxel is calcu- School of Engineering, the University of Vermont, Burlington, Vermont, lated by the volume fraction. It allows to analyze very large scale cal- United States culation rapidly. Correspondence: Hua Cao RESULTS To investigate the effect of phase controlling the simulation Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P17 using actual macaque monkey’s CT data is conducted. Figure 1 is simulation results and each image shows the pressure distribution OBJECTIVES High intensity focused ultrasound (HIFU) has become a surrounding the target point. Intensity of pressure is normalized by new noninvasive surgical modality for cancer treatment, however, maximum value of focal point. (a) shows the case of only waterme- the HIFU focusing precision is handicapped by the diffraction limit of dium and (b), (c) show the result under existence of bone without the wavelength of a traveling ultrasonic wave. A new HIFU trans- control and with control respectively. The focusing with control is im- ducer has been developed, which uses standing waves generated in proved compared to focusing without control. Figure 2 shows the re- a spherical cavity with open ends and break the diffraction limit to sult of 3 different ch number of array transducer. Compared with achieve subwavelength focusing region. 16ch, 64ch has better focusing but there is no large difference be- METHODS In order to describe the acoustic field, a finite element tween 64ch and 128ch. model is developed to numerically study the acoustic field gener- CONCLUSIONS Simulation result suggests that the phase controlling ated from the spherical cavity transducer assembly and the ex- can work effectively. And it is also founded that by comparing the re- periment was setup to measure the frequency dependence of sult of 3 different number ofch, 64ch is enough to control the focal the acoustic field generated by this spherical cavity transducer. point. In the future in order to verify the simulation result we are go- RESULTS Our theoretical and experimental results demonstrate ing to compare it with experimental measurement. Then parameters that in its resonant modes, the focusing zone is smaller while the and modeling method of simulation are reconsidered for the experi- focusing gain of sound pressure ishigher. ment using macaque monkey. CONCLUSIONS These results indicate that the focal zone of the acoustic field inside a spherical cavity with open ends is compressed to a sub-wavelength level while the intensity of sound pressure in the focal region significantly increases. P18 Non-invasive therapeutic method for brain disease using TFUS system assisted by numerical simulation 1 2, 1 3 3 Yohei Kobayashi , Takashi Azuma , Tatsuya Umeda , Tomomichi Oya , 3 4 1 4 Masashi Koizumi , Ryo Suzuki , Naoto Yamamura , Kazuo Maruyama , 3 1 Kazuhiko Seki , Shu Takagi 1 2 Bioengineering Dept., The University of Tokyo, Tokyo, Japan; Faculty of Medicine, The University of Tokyo, Tokyo, Japan; National Center of Neurology andPsychiatry, Kodiara-shi, Tokyo, Japan; Teikyo University, Tokyo, Japan Correspondence: Yohei Kobayashi Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P18 OBJECTIVES tFUS (transcranial focused ultrasound) has a great po- tential for non-invasive therapy for brain disease. The purpose of this research is to apply tFUS with microbubble for BBB opening for DDS and stimulation of red nucleus for motion trigger. BBB opening is the technique to enhance the permeability of blood brain barrier for effi- cient drug delivery. On the other hand, stimulation of red nucleus which exists in deep area of brain is said to be effective against re- habilitation from cerebralinfarction. For investigating these therapy, we conduct the experiment using macaque monkey which is closer to the human. In either case major issues concerning application of tFUS for brain therapy is focal displacement due to reflection and re- fraction through the skull. In this research we develop the focal con- trolling method with array transducer and simulation of ultrasound propagation utilizing CT data of macaque monkey. Fig. 1 (abstract P18). Pressure distribution surrounding target METHODS Array transducer can control the focal point by adjusting point(a) Water (b) skull without control (c) skull with control phase of each element. Phase delay of each element can be decided Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 67 of 122 harmonic responses grow with increasing acoustic pressure for both chirp and sinusoidal excitations and this growth would reach a satur- ation level when acoustic pressure exceeds a specific threshold.The ex- perimental results agree with simulations when the driving pressure is less than 300 kPa, but the saturation amplitudes of sub-harmonic re- sponses for both excitations after that are much lower than the simu- lated results. When the driving pressure is greater than 100 kPa and less than 300 kPa, the sub-harmonic amplitudes excited by chirp signals are always 5-10 dB higher than those excited by sinusoidal signals. The stable cavitation threshold for chirp excitation is also observed to be much lower than for sinusoidal excitation. Optimization: As the band- width of chirp signal increase, the sub-harmonic amplitudes changes more greatly when the ambient pressure changes. This indicates that the wider bandwidth could offer a better sensitivity in ambient pres- sure evaluation by exciting a wider size range of microbubbles. How- ever, bandwidths over 22.8% were not investigated because of the overlap effect wider bandwidths may bring. The studies on chirp pulse length indicates that the measured sub-harmonic responses increase with longer pulse length. The inherent reason for this behavior is that each individual microbubble may take a few cycles to reach a steady state when a strong non-linear effect is observed. CONCLUSIONS The sub-harmonic waves produced by stable cavita- tion of bubbles provides the feasibility of enhanced blood pressure measurement. Due the non-negligible size distribution of commercial ultrasound contrast agent microbubbles, chirp signals rather than si- nusoidal signals are applied in this article to excite a wider size range of microbubbles. Chirp ultrasound has been theoretically and experi- mentally verified to effectively enhance both the amplitude of sub- Fig. 2 (abstract P18). Pressure distribution surrounding target harmonic response and its sensitivity to ambient pressure, thus it pointof three different number of ch overweighs conventional sinusoidal signals in pressure measurement. Wider bandwidths and longer pulse lengths for chirp excitation also prove to realize an optimized pressure evaluation routine. Further P19 potential development of the proposed method will require en- Evaluation of ambient pressure using sub-harmonic response of hancement of bubble dynamics models and improved microbubble microbubbles sonicated with chirp pulses fabrication techniques so that this method could work better for Zhiyang Jin, Siyu Liu, Xiasheng Guo, Juan Tu, Dong Zhang non-invasive blood pressure measurement applications. Nanjing University, Nanjing, China Correspondence: Zhiyang Jin Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P19 OBJECTIVES The sub-harmonic signal generated by ultrasound contrast agents’ stable cavitation has been proven to be a potentially effective and efficient cue for noninvasive blood pressure measurement. In this work, an improved ambient pressure evaluation method is proposed to enhance the sub-harmonic responses of microbubbles so that a more sensitive and accurate pressure measurement could be achieved. METHODS Chirp signals, namely, linear frequency-modulated signals, are combined with microbubble sub-harmonics in this work. Both simulations and experiments are conducted to illustrate the advan- tage of chirp excitation in ambient pressure evaluation by compari- sons with conventional sinusoidal excitation. Dependence of subharmonic response on chirp parameters, namely, acoustic pres- sure, central frequency, bandwidth and pulse length, are also studied for optimization of sub-harmonic responses under chirp sonication. All the simulations are based on Marmottant model, supposing microbubbles have a Gaussian size distribution. Commercially availa- Fig. 1 (abstract P19). See text for description bleSonoVue microbubbles (Bracco Diagnostics, Geneva, Switzerland) were used for the experimental measurements and the experimental setup is illustrated in Fig. 1 uploaded. RESULTS Sub-harmonic enhancement by chirp excitation: SonoVue microbubbles are driven by sinusoidal and chirp excitation sharing the P20 same driving pressureamplitude and pulse length at Pov = 0 kPa (spe- Development of a focused ultrasound device for skin ablation 2 1 2 1 cifically, ambient pressure of 1 atm). The sinusoidal wave frequency Meng-Hung Tsai , Li-Chen Chiu , Win-Li Lin , Gin-Shin Chen and the central frequency of the chirp excitation werefs= fc = 3.5 MHz, Institute of Biomedical Engineering and Nanomedicine, National Health and the chirp signal bandwidth was Δf = 0.4 MHz. A significant increase Research Institutes; Institute of Biomedical Engineering, NationalTaiwan of 5.1 dB in the sub-harmonic component can be observed in the chirp University excitation case. As the ambient pressure increases, the measured sub- Correspondence: Meng-Hung Tsai harmonic responses induced by chirp excitation reduce more violently Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P20 than sinusoidal case, providing better sensitivity in ambient pressure measurement. All the experimental results agree well with the OBJECTIVES Focused ultrasound can concentrate acoustic power on simulations. Dependence of sub-harmonic response on driving acoustic the target tissues like subcutaneous fat and superficial muscular apo- pressure: In a similar setup with the former one, the amplitudes of sub- neurotic system to generate a steep temperature elevation, leading to Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 68 of 122 collagen denaturation, contraction and remodeling. In the study, a sin- gle-element focused ultrasound transducer integrated with an auto- matic motion system was developed for multi-point ablation in the skin. METHODS The transducer was made of 1-3 piezocomposite mate- rials with a diameter of 10 mm and a radius of curvature of 17 mm (Fig. 1). The operation frequency was 5MHz. The electro- acoustic conversion efficiency and focal zone were measured after electrical impedance matching. The effectiveness of the de- vice was evaluated by the ablation experiments of phantom, ex- vivo and in vivo rat model. RESULTS The efficiency of the transducer was 43.9±2.6%. The focal depth and focal width were 7.2 and 0.7 mm, respectively. With the ultrasonic parameters of electrical power of 10 W for 1 s, the lesions were formed in the 52°C thermo-sensitive hydrogel phantom (Fig. 2). Fig. 3 (abstract P20). The H&E staining of the rat skin tissue after Ex vivo and in vivo studies showed that the transducer could pro- ultrasonic sonications duce a hot spot to noninvasively cause superficial skin necrosis as the electrical power and time of ultrasonic sonication were in a range of 6-9 W and 2 s. CONCLUSIONS A relatively high-frequency focused ultrasound device P21 has been developed for skin ablation. In vitro and animal studies Low intensity pulsed ultrasound stimulates hair follicle cells in 3D have verified the efficacy ofthe ultrasonic device (Fig. 3). culture Noboru Sasaki, Mitsuyoshi Takiguchi Hokkaido University, Sapporo, Japan Correspondence: Noboru Sasaki Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P21 OBJECTIVES Low intensity pulsed ultrasound (LIPUS) has been known to activate protein synthesis, DNA synthesis and cell prolifera- tion in several different types of cells. This proof of concept study evaluated whether LIPUS stimulates proliferation of hair follicle cells in 3D culture. METHODS The 3D culture was composed of three layers. In the top layer, Human Follicle Dermal Papilla cells (HFDPC; PromoCell) were embedded into Matrigel (Corning). The middle layer consisted of only Matrigel. In the bottom layer, Human Hair Outer Root Sheath cells (HHORSC; ScienCell) were embedded into Matigel.Low intensity pulsed ultrasound was exposed to cells from above the 3D culture. Ultrasound parameters are as follows; 1 MHz center frequency, 500 pulses, 1 kHz PRF (i.e. the duty factor was 50%), Isata 90 mW/cm2. After ultrasound exposure, cells in the 3D culture were stained by Calcein-AM and observed by a fluorescent microscopy. RESULTS Both HFDPC and HHORSC were increased by single expos- ure of LIPUS. Moreover, HFDPC grew upward, i.e. from the bottom Fig. 1 (abstract P20). The photo of the 5MHz focused layer to the top layer. ultrasonic transducer CONCLUSIONS This study showed the feasibility of LIPUS for stimu- lating the proliferation of hair follicle cells. Further in depth study may assess mechanisms of thestimulation and optimize ultrasound parameters. P22 MR thermometry of a novel focused ultrasound application in post-mortem skin 1 2 Ziqi Wu , Luis Hernandez-Garcia 1 2 Access Business Group, Grand Rapids, Michigan, USA; Biomedical Engineering, University of Michigan, Ann Arbor, Michigan, USA Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P22 OBJECTIVES Intense focused ultrasound has been introduced in past years for non-invasive anti-aging facial treatment. By creating micro- necrosis in dermalor superficial muscular aponeurotic system (SMAS), wound healing process can be triggered followed by promotion of fi- broblasts activity and collagenproduction. Recent research showed that repeated mild application of focused ultrasound energy in the skin with lower acoustic intensity also provided clinical benefits. Temperature rise generated inside the skin using this technique is hypothesized as the mechanism of action, and the estimation of temperature within the targeted region is important for the applica- Fig. 2 (abstract P20). A white lesion was induced in the tion efficacy and safety. In this study, we demonstrated that MR thermalsensitive hydrogel phantom by the developed transducer thermometry can be used to measure the 2Dtemperature changes Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 69 of 122 generated by a novel focused ultrasound skin applicator with fine spatiotemporal resolutions in post-mortem skin. METHODS A MR compatible prototype transducer was constructed (Ac- cess Business Group, USA) using a partial cylindrical ceramic element and a plastic waveguide. The focal distance was confirmed to be 4 mm in water by Schlieren imaging and the -6dB focal region was measured to be 1mm inwidth and20mmin lengthusing ascanninghydrophone system. The maximum rarefraction pressure was less than 1 MPa while the spatial-peak pulse-average intensity was less than 30 W/cm2. The transducer was driven with a function generator and an amplifier at a center frequency of 4.5 MHz with 2.8 W average acoustic power. Pulsed ultrasound (400 ms) was generated with a duty cycle of 89% and total Fig. 2 (abstract P22). (A) MR image acquired using gems sequence. exposure time varied from 7 to 30 seconds. A 7T MR scanner (Agilent- Postmortem full thickness skin is roughly 1 cm thick overlaid on top Technologies, Walnut Creek, CA) was used to image porcine samples for of a piece of porcine muscle which helpedavoiding reflection from repeatability of MR thermometry and followed by imaging the post-mor- the bottom of the chamber. (B) Trace of the maximum temperature tem skin. Highresolution MR images were acquired with multislice gradi- rise in the skin from a 30 second ultrasound exposure ent-echo sequence (TR/TE = 20/4 ms, flip angle = 20, voxel size = 0.47 x 0.47 x 2 mm) at the beginning and end of the experiment while gradient echo EPI sequence (TR/TE = 400/11.32 ms, triple shots, voxel size = 0.42 x 0.83 x 1 mm) were performed before, during, and after ultrasound expos- P23 ure. The ultrasound transducer and the MRI scanner were synchronized Ultrasound-mediated transdermal delivery of hyaluronic acid into via TTL pulses, such that the EPI images were collected between ultra- skin sound bursts. Spatiotemporal temperature changes were computed Yi Yun using proton resonant frequency shift relationship from the MR phase VITA-Sound Tech, USA images. Finally, skin samples were stored in 10% formalin, fixed in paraf- Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P23 fin, sliced, and stained with H&E and masson’s trichrome to investigate thermal damage on skin cells. Transdermal drug delivery (TDD) can effectively bypass the first-pass RESULTS The reconstructed temperature images in the coronal plane effect, which can be valuable in cosmetic industry. In our work, across the ultrasound beam width showed the best image quality (Fig. 1) ultrasound-facilitated TDD on fresh porcine skin was studied in vari- whereas the sagittal plane images along the beam length were poor due ous conditions of acoustic parameters. The delivery of fluorescent to magnetic field inhomogeneity. In the porcine muscle, maximum nanoparticles and high molecular weight hyaluronic acid (HA) in the temperature rise of 13.2 ± 0.3 °Coccurred inside the focal region and the skin samples was observed by laser confocal microscopy and ultra- MR thermometry results were repeatable with four subsequent ultra- violet spectrometry, respectively. The results showed that, with the sound exposures. Transient changes were also observed on MR magni- application of ultrasound exposures, the permeability of the skin to tude images. In the post mortem skin, 8 seconds ultrasound exposure these markers (e.g., their penetration depth and concentration) could generated temperature rise of 17.9 °C whereas 30 seconds exposure be obviously raised above its passive diffusion permeability. More- caused 33.3 °C temperature increase (Fig. 2). The maximum temperature over, ultrasound-facilitated TDD was also tested with/without the rise occurred at 2 mm from the transducer surface into the dermis. No presence of ultrasound contrast agents (UCAs). When the ultrasound permanent histological cell damage was seen for shorter ultrasound ex- was applied without UCAs, low ultrasound frequency will give better posure whereas necrosis was observed with longer ultrasound exposure. drug delivery effect than high frequency, but the penetration depth CONCLUSIONS Using gradient echo EPI sequence, MR can measure was in a less level around 200 μm. However, with the help of the spatiotemporal temperature changes induced by a novel ultra- ultrasound-induced microbubble cavitation effect, both the penetra- sound skin applicator inpost-mortem skin and provide potential tion depth and concentration in the skin were significantly enhanced safety temperature or thermal dose threshold for focused ultrasound even more. The best ultrasound-facilitated TDD could be achieved dermatological applications. (A) Reconstructed temperature maps with a drug penetration depth of over 600 μm, and the penetration based on PRF shift superimposed with high resolution gem image; concentrations of fluorescent nanoparticles and HA increased up to (B) Traces of the maximum temperature. about 4-5 folds. In order to get better understanding of ultrasound- facilitated TDD, scanning electron microscopy was used to examine the surface morphology of skin samples, which showed that the skin structure changed greatly under the treatment of ultrasound and UCA. The present work suggests that, for TDD applications (e.g., nanoparticle drug carriers, transdermal patches and cosmetics) in cosmetic industry, protocols and methods presented had shown us the potentially attractive application for moisture and treatment of Skin Deapth. P24 Comparative study on ultrasonic monitoring of pHIFU and cHIFU peripheral ablation mode Wen Jing Chongqing Medical University, Chongqing, China Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P24 Fig. 1 (abstract P22). (A) Reconstructed temperature maps based OBJECTIVES To investigate the feasibility of Ultrasonic monitoring on on PRF shift superimposed with high resolution gem image; (B) PHIFU and CHIFU peripheral ablation mode. Traces of the maximum temperature rise measured in porcine METHODS 60 cases ox-liver tissues were divided into group A muscle from 4 subsequent ultrasound exposures (PHIFU, DC=15%, n=30) and group B(CHIFU,DC=100%,n=30).Under Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 70 of 122 the guidance of B mode ultrasound, each group was performed with P26 peripheral ablation, during and after the ablation, capture images of Concentration of MSNC-PFP influence surface models in high-intensity every single line and layer related to the target tissue, and then focused ultrasound ablation of ex vitro bovine liver analyze the gray scale change. After the whole tissue peripheral abla- Ding Xiaoya, Dazhao Ma, Wen Jing, Qi Wang, Jianzhong Zou tion, slice the target and observe its damage situation. Chongqing medical university, Chongqing, China RESULTS In the course of peripheral ablation, both groups were seen in- Correspondence: Ding Xiaoya stant echo enhancement in the irradiated areas, with time varying, the Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P26 Hyperecho were finally replaced by hypoecho in group A and the tissues were presented liquefaction necrosis, while linear Hyperecho were still OBJECTIVES To Discusse the influence of perfluoropentane drops-en- found in group B and tissues were presented coagulation necrosis. After capsulated mesoporous silica nanocapsules (MSNc-PFP) injected into peripheral ablation, a cyclic hypo-echo were observed around the target the melting region withthe surface ablation of HIFU in vitro bovine and the gray scale of the internally areas were slightly changed and tis- liver. sue observing nearly showed damaged. While in group B, the instant METHODS Ox-liver tissues were randomly divided into 5 groups: PBS Ultrasonic monitoring images were not agreed with the form of the tis- blank control group, 0.25 mg/ml group, 0.5 mg/ml group, 1 mg/ml sue damage situation for the entire target were presented time group, 2 mg/ml group according to the concentration of MSNc-PFP, depended strong echo change, and the areas where there is actually co- melting line isometric injection five points, measured 100 uL. Each agulation necrosis were partial hypoecho or medium echo. group was surface ablationed that the edge internal were not in- CONCLUSIONS The tissue damage character, ultrasonogram and gray cluded with the same dose of HIFU energy after rejection. scale change were all varied between PHIFU and CHIFU peripheral RESULTS 1 mg/ml group, 2 mg/ml group, the high-level echo of syn- ablation mode. It is possible to apply Ultrasonic monitoring while ergistic agent in the injection area was receding after injection; After performing peripheral ablation mode, but there is still limitation. HIFU irradiation, regional gray values were heighten and the scope was broadening changing with the concentration of synergist in- creased. Each group can form a complete coagulation necrosisexcept P25 the control group. And the higher the synergistic agent concentra- Experimental study of the effect of the target blood vessels angled tion used, the width and the ablation range of oagulation necrosis with acoustic axis on the surface ablation of pHIFU were greater. 2 mg/ml group showed obvious thermal damage per- Yang S. Ying, Zou Jianzhong ipheral ablation region. Imaging and Nuclear Medicine, Institute of biomedical engineering, CONCLUSIONS Concentration of 1 mg/ml of MSNc-PFP had a good Yuzhong district, Chongqing, China effect on the surface models of synergistic HIFU ablation. Correspondence: Yang S. Ying Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P25 P27 OBJECTIVES To explore the effect of the target blood vessels angled Effects of subatmospheric pressure and dissolved oxygen with acoustic axis on the formation of the around coagulation necro- concentration on the generation of lesions in ex vivo bovine livers sis under the model of the surface ablation of PHIFU with the same by HIFU conditions of irradiation dose, To adjust the HIFU treatment model Min He, Zhiqiang Zhong, Xiaobo Gong, Faqi Li, Deping Zeng, has certain guiding significance. Zhibiao Wang METHODS Embedded the rabbit thoracic aorta (diameter was 4.25 College of Biomedical Engineering, Chongqing Medical University, ±0.79 mm) into egg white body model and divided into 3 groups by Chongqing, China the blood vessels Angled 0 °,45°,90 °with acoustic axis.The egg white Correspondence: Min He body model were exposed to pHIFU with 200-300 w acoustic Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P27 power,100Hz pulse recurrence frequency(PRF),50% dutycycles.The surface ablation size is 30 * 30 * 30 mm3,line scanning, Linear speed: OBJECTIVES This study was aim to investigate the effects of subat- 3 mm/s;Line length: 30 mm; line scanning time interval: 1.5 min;Layer mospheric pressure and dissolved oxygen concentration on the spacing 2 mmand 16 layers in total.In the process of irradiation ultra- morphology and size of lesionsgenerated by HIFU in ex vivo bovine sonographic observations and temperature measurement; After the liver. irradiation, cut body model to observe theirradiation damage. METHODS All HIFU experiments were conducted in a stainless cham- RESULTS In the process of Irradiation, near the acoustic source side of ber fulfilled with degassed water. A 1MHz HIFU transducer was used the target blood vessels Angled with acoustic axis, the temperature to generate the US exposureof 11700W/cm2 which was acutely cavi- curve shown as: increase fast, long duration and falling fast and far tation under atmospheric pressure. The dissolved oxygen concentra- from the acoustic source side one is relatively slow, short duration, de- tion (DOC) of degassed water were divided into three groups: 1.0 creased slower. And the ultrasonographic performance that far from mg/L, 1.5 mg/L and 2.0 mg/L respectively. The ex vivo bovine livers the acoustic source side is low echo immediately, gradually enhanced, were exposing 2 seconds per time under two ambient pressure of at- after 3 min the gray level change is not obvious and near the acoustic mospheric pressure and subatmospheric pressure. B mode US moni- source side is immediatestrong echo, gradually weakened and after 3 toring the strong echo signal before and after HIFU exposing. A min the gray does not change significantly.After irradiation, cut down passive cavitation test system (PCD) test the acousticcavitation signal the egg white body models,observed that far from theacoustic source in the process of exposing. side of the blood vessel Angled 45 °, 90° with acoustic axis were no RESULTS 1. The broadband noise of atmospheric pressure and subat- white coagulation necrosis formed. And the 0°one coagulation necrosis mospheric pressure under the same dissolved oxygen concentration around the blood vessels was narrowed. shows that cavitation was weaker under subatmospheric pressure CONCLUSIONS Adopted the surface ablation model, the existence of than that under atmospheric pressure.2. The variation of difference the target blood vessels angled with acoustic axis can affects its sur- of gray level value between before and after HIFU exposing on the rounding coagulation necrosisformation. The target blood vessels An- B-scan image is increasing with the increase of dissolved oxygen gled 0 °with acoustic axis can narrowed the coagulation necrosis concentration. Under atmospheric pressure, the difference of gray arround the vessels. And far from the acoustic source side of the- level value is larger than that under subatmopheric pressure.3. Under blood vessels Angled 45 °,90 °could not formed the coagulation ne- atmospheric pressure, the lesions are from tadpole shape to elliptical crosis. And in the process of Irradiation, the change of the gray-scale as the dissolved oxygen concentration decreased. But under subat- ultrasonography could judge thedamage formation around the blood mopheric pressure, thelesions are all elliptical. The size of lesions is vessels to a certain extent. increasing with the increase of dissolved oxygen concentration. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 71 of 122 Under the same dissolved oxygen concentration, the size oflesions is RESULTS Effective follow-up were performed in 297 cases. The fol- larger under atmospheric pressure than that of subatmospheric low-up time was 1—50 months. There were 177 cases in diffuse pressure. group, while 120 cases in focalgroup. The incidence of ablation CONCLUSIONS This study investigated the effects of subatmospheric among 297 patients was 99.33% (295/297). The NPVR of diffuse and pressure and dissolved oxygen concentration on the generation of focal group were ([26.00±13.36] %) and ([44.32±19.93] %), respect- lesions in ex vivo bovine liversby HIFU. The followings were clarified ively. The dysmenorrhea and menorrhagia score of post-procedure in the experiment.1. The reduce of dissolved oxygen concentration were significantly lower than those of pre-procedure in two groups could decrease the volume of lesions generated by HIFU.2. Subatmo- (both P<0.05). The total remission rate of dysmenorrhea in 3, 6, 12, spheric pressure could restrain the cavitation, thus reshape the le- 24 and 36 months after ablation were 92.96% (264/284), 86.18% sions to elliptical and smaller the size of lesion in focus. (237/275), 73.51% (197/268),60.71%(136/224), 46.83% (59/126), re- spectively. The remission rate of dysmenorrhea of focal group was higher than that of diffuse group in each follow-up period, while sig- P28 nificant differences were observed in 6, 24 and 36 months (P<0.05). The effect of phased-hifu with discontinuous operating mode on The total remission rate of menorrhagia in 3, 6, 12, 24 and 36 months coagulative necrosis region after ablationwere 87.38% (187/214), 83.09% (172/207), 68.63% (140/ Xiongfei Qu, Guofeng Shen, Nan Wu, Yazhu Chen 204), 63.64% (105/165), 45.92% (45/98), respectively. The remission School of Biomedical Engineering, Shanghai Jiao Tong University, rate of menorrhagia of focal group washigher than that of diffuse Shanghai, China group in each follow-up period, while no significant difference was Correspondence: Xiongfei Qu observed between two groups (all P>0.05). Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P28 CONCLUSIONS HIFU is significant effective for adenomyosis, the short-term efficacy of focal and diffuse type are familiar, while the OBJECTIVES In the previous experiments, we noticed that Phased- long-term efficacy of focal typeis better than that of diffuse type. HIFU with discontinuous operating mode (eg. 2s heating following with a 1s cooling, and repeating) produced shorter and thicker co- agulative necrosis region in ex vivo porcine muscle, compared with slender spindle like region in continuous operating mode. The aim of P30 this study was to demonstrate the mechanism and influence of this Ultrasound-guided versus mr-guided high intensity focused method on tissue ablation. ultrasound for ablation of uterine fibroids by ultrasonic contrast METHODS In this study, we investigated the influence of discontinu- agent: treatment efficacy, safety and efficiency ous operating mode on tissue ablation. Three different treatment Yi Wang, Yonghua Xu procedures in 4 repeat cycle (each for 2.35s) were simulated using a Chongqing Medical University, The Institute of Ultrasound Engineering DFDT method: (1) Continuous heating with constant 200 watts of in Medicine, Chongqing, China acoustic power; (2)50% duty ratio (1.175s heating in a cycle) heating Correspondence: Yi Wang with 200 watts of acoustic power; (3)50% duty ratio heating with 400 Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P30 watts of acoustic power. Then the ex vivo porcine muscle and tissue- mimicking phantom (NIPAMbased hydrogel phantom with cloud OBJECTIVES The time efficiency and sonication energy efficiency point temperatures at 52°C) heating experiments were performed in of treatment and safety were compared between MR-guided procedures (1) and (3), to investigate the shape of coagulative necro- High-Intensity Focused Ultrasound (MRgHIFU) and Ultrasound- sis region and temperature above 52°C, respectively. guided High Intensity Focused Ultrasound (USgHIFU) for complete RESULTS The simulation showed that the short rod-like 240EM re- ablation of T2 hypointense fibroids. gion of the discontinuous operating mode was shorter, thicker and METHODS The treatment data and sonication parameters from 13 larger, compared with slender spindle like 240EM region of continu- uterine fibroids in 10 patients treated with MRgHIFU and 28 uter- ous operating mode. However, the gradient of thermal dose in ine fibroids in 22 patients treated with USgHIFU were analyzed. 240EM region boundary of discontinuous operating mode was much All of the pre-treatment fibroids were hypointense signal in T2 smaller, which may cause the instability of ablation border. These weighted imaging and completely ablated by using MRgHIFU two simulation results were observed in the ex vivo porcine muscle orUSgHIFU at one session treatment. The volume of the treated and tissue-mimicking phantom experiments. fibroid and the non-perfused volume (NPV) was calculated in CONCLUSIONS The discontinuous operating mode can produce a lar- contrast enhance MRI (CE-MRI), andcomplete ablation of fibroids ger short rod-like coagulative necrosis region, therefore may reduce is defined as non-perfusion region covering all volume of the the number of treatment shotsand improve the efficiency of treatment. treated fibroid immediately following the procedure. The treat- However, it reduces the gradient of thermal dose in ablation boundary, ment and sonication time, the EEF and NPV ratios were com- therefore may reduce the stability of coagulative necrosis region. pared between MRgHIFU and USgHIFU, while the adverse events and complications were also assessed. RESULTS The percentage rates of the completely ablated fibroids in P29 the MRgHIFU and USgHIFU were 29.5% and 41.2%, respectively. The Short-term and long-term efficacy of ultrasound ablation for treatment time was174.5±42.2 minutes and 114.4±39.2 minutes, the diffuse and focal adenomyosis sonication time was 24.7±9.0 minutes and 19.4±6.8 minutes, the son- Yujie Feng, Jinyun Chen ication power was 310.2±62.5W and 391.6±16.6W, the sonication en- College of Biomedical Engineering, Chongqing Medical University, ergy was 483.0±248.2 KJ and 463.2±156.4KJ, EEF was 5.1±3.0 KJ/cm3 Chongqing, China and 6.8±5.2 KJ/cm3, and the mean treatment speed was 42.2±25.6 Correspondence: Yujie Feng cm3/h and70.9±41.9 cm3/h in the MRgHIFU and USgHIFU treatment Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P29 for complete ablation of fibroids, respectively. There was a negative linear correlation between the EEF andthe NPV of fibroids, and a OBJECTIVES To compare the short-term and long-term efficacy of HIFU positive linear correlation between the treatment speed and the NPV in treatment of diffuse and focal adenomyosis. of fibroids in the both groups (P<0.05). There was a positive linear METHODS A total of 308 patients with adenomyosis who accepted correlation between the sonication intensity and the NPV of fibroids HIFU ablation were collected. According to preprocedural MRI, the in the USgHIFU group (P<0.05) and no correlation in the MRgHIFU patients were divided into diffuse and focal group. The non-perfused group (P>0.05). There was no severe adverse event and major com- volume ratio (NPVR) and the incidence of ablation were calculated. plication in both groups after treatment. Preprocedural and postprocedural situation of dysmenorrhea and CONCLUSIONS MRgHIFU and USgHIFU both are feasible, safe, and ef- menorrhagia were evaluated. fective with the equivalent energy efficiency for complete ablation of Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 72 of 122 T2 hypointense fibroids;USgHIFU was superior to MRgHIFU in the [3] Younan et al. "Influence of the pressure field distribution in transcranial time efficiency. ultrasonic neurostimulation." Medical physics (2013) [4] Ye et al. "Frequency Dependence of Ultrasound Neurostimulation in the Mouse Brain." Ultrasound in medicine & biology (2016) P31 [5] Li et al. "Improved Anatomical Specificity of Non-invasive Neuro- Estimation of thermal rise during ultrasonic neurostimulation in stimulation by High Frequency (5 MHz) Ultrasound." Scientific reports rodents: retrospective analysis of five recent studies (2016). Charlotte Constans, Mickael Tanter, Jean-Francois Aubry [6] Yoo et al. "Transcranial focused ultrasound to the thalamus alters Institut Langevin, Paris, France anesthesia time in rats." Neuroreport (2011) Correspondence: Charlotte Constans [7] Kamimura et al. "Focused ultrasound neuromodulation of cortical and Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P31 subcortical brain structures using 1.9 MHz." Medical Physics (2016) [8] B. Cox et al, k-space propagation models for acoustically heterogeneous OBJECTIVES The first ultrasonic neuromodulation studies were con- media: Application to biomedical photoacoustics, J. Acoust. Soc. Am., ducted with rather high intensity [1] and thermal effects were as- 2007. sumed to be the main cause [2]. More recently, multiple groups have [9] Burgoon et al. "Temperature-sensitive properties of rat suprachiasmatic reported successful low intensity focused ultrasound (LIFU) neurosti- nucleus neurons." American Journal of Physiology-Regulatory, Integrative mulation on rodents: movement elicitations [3] [4] [5] [7] or reduction and ComparativePhysiology (2001) of anesthesia time [6]. Given the low intensities used in most of them, mechanical effects are more prone to induce neuromodulation Table 1 (abstract P31). See text for description [4]. The mechanism of neuromodulation is still not fully understood, and a more thorough study of the thermal and mechanical effects is necessary to optimize the parameters for clinical applications. We simulated the thermal rise in 5 rodent studies in order to evaluate its potential impact. METHODS The acoustic propagation of focused ultrasound was simu- lated in an entire rat head in order to investigate the pressure ampli- tude and spatial distribution. The simulations were performed with k- Wave [6], a k-space pseudospectral method-based solver. 3D maps of the skull, brain and tissues were extracted from a rat microcomputed tomography scan. Brain and tissues were assumed to have the same sound-speed and density as water, and the transducer was modeled according to each study’s materials. Absorption was taken into ac- count in the skull (2.7dB/cm/MHz) and in the brain (0.21dB/cm/MHz) with a 1.18 power law of frequency. Ultrasound propagate in a cone filled with water before entering the rat head, the geometrical focal point being located about 7mm deep from the surface, inside the brain. For each study, we calculated the pressure at focal spot in water based on each study materials and methods. The simulations were first performed in water and compared to these extracted am- plitudes. The peak negative pressure in the rat head was extracted from the simulation and thus takes into account reflections and ab- sorption effects. The thermal code is based on the bio-heat equation without perfusion and metabolic processes. The thermal dose unit is Fig. 1 (abstract P31). Maximum temperature in brain, with zoom CEM (cumulative equivalent minutes at 43°C). on individual bursts (left) and at focal spot (right) in study [3] RESULTS Parameters and results in brain and at the focal spot for all the studies are listed in Table 1. Temperature rise estimated for [3] and [7] are plotted on Figs. 1and 2 respectively for the most heated point in the brain (left) and at the focal spot (right). In study [3], the thermal does not exceed 3.3E-4 CEM in the skull, brain and skin. In study [7], TD reaches 2800 CEM in the skull, 60 CEM in the brain (close to thebone) and 50 CEM in the skin. The thermal dose in the skin is significant but not high enough to induce skin burns. CONCLUSIONS Our retrospective analysis shows thermal rise ranging from 0.002°C to 9.3°C in the brain. For studies [3-6], corresponding to a frequency range of320kHz to 5MHz and a total sonication time ran- ging from 80ms to 20min, the maximum temperature elevation in the rodent brain is lower than 0.1°C. Sensitivity to temperature changes was found with a coefficient of 1.1 impulses/s/°C in some neurons [9]. Thus, in the case of study [7], the thermal rise cannot be Fig. 2 (abstract P31). Maximum temperature in brain, with zoom neglected as apossible cause of neuromodulation. on individual bursts (left) and at focal spot (right) in study [7] Acknowledgements This work was supported by the Bettencourt Schueller Foundation and the "Agence Nationale de la Recherche" under the program “Future Investments” P32 with the reference ANR-10-EQPX-15. Numerical simulation of the effect of phase transformation on standing waves and transcranial focusing in HIFU 1 1 2 1 References Miaomiao Zeng , Shihui Chang , Rui Cao , Xiqi Jian [1] Fry et al. "Production of reversible changes in the central nervous system Biomedical engineering, Tianjin Medical University, Tianjin, China; by ultrasound." Science (1958) Tianjin University of Science and Technology, Tianjin, China [2] Lele et al. "Effects of focused ultrasonic radiation on peripheral nerve, Correspondence: Miaomiao Zeng with observations on local heating." Experimental Neurology (1963) Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P32 Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 73 of 122 OBJECTIVES Reflections induced by heterogeneous structures and P33 large differences in acoustic impedance between the skull and peri- Analysis and investigation of the major parameter effecting enchyma result in undesired standing waves that cause energy loss transcranial ultrasound phase aberrations: a preliminary study in the treatment area in high-intensity focused ultrasound (HIFU) Nan Wu, Guofeng Shen, Xiongfei Qu, Yazhu Chen transcranial treatment and deposition of excess energy in healthy tis- School of Biomedical Engineering, Shanghai Jiao Tong University, sue. The goal of this work is to address these issues. Shanghai, China METHODS The simulations performed in this paper were based Correspondence: Nan Wu on the three dimensional finite difference time domain (FDTD) Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P33 simulation of acoustic propagation and thermal behavior through the bone windows. The material properties of the skull were de- OBJECTIVES Focused Ultrasound (FUS) is a noninvasive medical rived from 3D reconstructions of high-resolution computed tom- technology used for transcranial therapeutic applications. How- ography (CT) scans of selected patients. Phase transformation ever, due to the complex acoustic properties of the skull, it is methods were used to reduce the standing wave by randomly practically hard to obtain a sharp transcranial focus without changing the phase in time segments. phase correction. The aim of this research was to analyze and in- RESULTS The intensity of the standing wave decreased (Fig. 1). vestigate the major parameter of the skull which has the greatest Meanwhile, the sound pressure rose and the rate of temperature rise effect on phase aberrations when the ultrasound propagates at focal region increased when using phase transformation. Different through the skull. The correction of this parameter may simplify bone windows exhibit different optimum excitation frequencies, in the method of ultrasonic phase correction. the range of 0.6-0.8MHz. The minimum standing wave intensities ap- METHODS The numerical model was based on the k-wave simula- peared when the ratio of the phase transformation frequency to the tion environment, which was extensively tested and actually be- excitation frequency was 0.3 for all bone windows (Fig. 2). ing used to simulate the ultrasound field before. The whole CONCLUSIONS The phase transformation method has been proved simulation environment was designed in water, and a digitized to be effective in suppressing standing waves through variety bone human skull profile, which was built from CT (computed tomog- windows. The advantage of this method is that it can enhance the raphy) images, was placed below the transducer. To investigate energy of focal region and reduce the intensity of standing waves. how each parameter of the skull effects phase aberrations, dens- The optimum excitation frequencies selected for different bone win- ity, attenuation, velocity and geometry of the skull were taken dows were obtained and it was confirmed that the excitation fre- into account individually. The wave propagation simulations were quency and phase transformation frequency were relevant. performed with one of the skull parameters, hypothesizing the others set to be the parameters of water throughout the simula- tion process. The benchmark configuration for this research was the spherically curved transducer driven at 700 kHz; the trans- ducer had a curvature radius of 120 mm and a diameter of 90 mm. The focus acquired from different parameters was compared with each other after the simulations. RESULTS A transducer placed in a homogeneous media (water), without and with the skull were simulated to be comparison groups. The focus of the transducer was shifted and defocused when the skull was placed in the sound field. However, a sharp focus still could be achieved when the density or attenuation of the skull was taken into account. An aberrant focus was gener- ated when the velocity was set to 2850m/s (velocity of the skull). It was interesting to note that, the more velocity vectors penetrate the skull perpendicularly, the better a focus could be Fig. 1 (abstract P32). The acoustic pressure distribution (temporal obtained. bone window, 0.6 MHz, I =4.0 Wcm , t =16 s). (a) without and (b) CONCLUSIONS This research presented a simulation to analyze with standing wave suppression and investigate the major skull parameter effecting transcranial ultrasound wave phase aberrations. The velocity of skull could be the major parameter on phase aberrations, compared with the density and attenuation. Besides, the more velocity vectors per- pendicular to the skull, the less phase aberrations and higher soundpressure couldbeobtained. Amethodonmakingmore velocity vectors penetrating the skull perpendicularly that can simplify the correction of ultrasound wave phase will be done in the further research. P34 Protective effect of ultrasound on brain injury in mice Feng-Yi Yang, Wei-Shen Su Department of Biomedical Imaging and Radiological Sciences, National Yang-Ming University, Taipei, Taiwan Correspondence: Feng-Yi Yang Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P34 OBJECTIVES The purpose of this study was to investigate the effects of low-intensity pulsed ultrasound (LIPUS) in mice with traumatic brain injury (TBI). METHODS Controlled cortical impact (CCI) injury was used as a TBI animal model. Mice subjected to CCI injury were treated with LIPUS Fig. 2 (abstract P32). Plot of Rα against fr./f (temporal scales area at daily for a period of 28days. Behavioral assessments (mNSS and 0.6MHz, occipital area and parietal area at 0.7MHz) rotarod) were performed at day 28 after TBI. Histological examination Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 74 of 122 was performed via cresyl violet staining. Edema regions were moni- normalized therapeutic gain plots for the lateral and elevation direc- tored by magnetic resonance imaging. tions. The hydrophone measured pre-focal planes showed strong RESULTS Our data showed that functional impairments were signifi- agreement with back propagated simulations. As the frequency is in- cantly improved by LIPUS stimulation at day 28 after TBI. LIPUS sig- creased, the distortion and attenuation increases. Figure 2 shows the nificantly preserved brain tissue compared with the non-treated TBI back propagation results for a single frequency of 3.2 MHz. When group. Furthermore, LIPUS significantly reduced T2-weighted lesion analyzed as a whole frequency set, they show sections of the skull hav- volume in injured mice compared with the non-treatedTBI group. ing different frequency propagation characteristics. Complex skull-bone CONCLUSIONS In summary, LIPUS stimulation improved long-term structures and suture lines increase focal distortion and decrease total behavioral outcomes and attenuated brain tissue damage in mice powertransmission. Contiguous regions display greater transmission subjected to TBI. Therefore, transcranial LIPUS stimulation may pro- with decreased distortion and foci shifting. Figure 3 shows the bregma vide a potential treatment modality for TBI. suture line (Plane 1) causes stronger distortions than transmission re- gions between the bregma and lambda suture lines. The grid of points behind the skull show the distortion is not constant throughout theb- rain. These distortions are dependent upon both frequency and the re- P35 gion of the skull the beam is traveling through to get to the respective Broadband characterization of focused ultrasound transskull imaging pixel. These results point towards creating a pre-treatment cal- transmission culation of the effectiveness of a successful therapeutic delivery. Parker D. O'Brien, Dalong Liu, Emad S. Ebbini CONCLUSIONS The results demonstrate the feasibility of DMUA im- Electrical and Computer Engineering, University of Minnesota–Twin aging (both SA and STF) to characterize the transmission loss Cities, Richfield, Minnesota, USA through the skull as can be seen from the general agreement with Correspondence: Parker D. O'Brien the hydrophone estimates (Fig. 3). The results also demonstrate that Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P35 a relatively wide transmission bandwidth in both rodent and human skulls. Based on these results, it would be possible to transmit broad- OBJECTIVES The future of transcranial Focused Ultrasound (tFUS) for band tFUS beams that can be designed to minimize the focal spot subtherapeutic (neuromodulation) or ablative treatments in the hu- and improve the specificity of both ablative and subtherapeutic (neu- man brain relies on spatially specific therapeutic delivery with quanti- romodulation) tFUS beams. fiable power control to enable reversible and irreversible treatments. Single frequency transmission signals may not provide adequately defined foci nor minimal foci shift after propagating through the varying complex structures of the skull to target fine neuro-structures within the brain. Broadband transmission is highly likely to provide the most effective method of delivering specialized therapeutic treat- ments obtainted through its ability to recover distortion and loss from a single frequency with a wide range of available frequencies. This paper presents the broadband transmission characterization of FUS through rat skulls and human skulls ex vivo through various re- gions of the skulls to understand how different frequencies can be used to refocus distorted or recover lost transmission through the use of multiple frequencies. METHODS Two dual-mode ultrasound arrays (DMUA) are used to transmit FUS through a series of human and rodent skulls. Both DMUAs are concave spherically focused, linear arrays with radii of curvature specific for the skull model used (r=100mm for human, r=40mm for rat) (Imasonic, France). Planar acoustic pressure mea- surements using needle hydrophones (Onda Corp, Sunnyvale, CA) were performed in degassed water with and without skull samples present (Fig. 1). Field scans were performed in prefocal and focal planes on a finely sampled grid with sub-wavelength spacing. Figure 1 shows the setup using a human skull in the water bath with the 600-micron hydrophone and DMUA (100-mm ROC) operating in the 0.7 - 2.5 MHz frequency range. Total power was estimated by inte- grating the square intensity (from hydrophone measurements). In Fig. 1 (abstract P35). See text for description addition, the focal plane measurements were back propagated to the interior of the skull to predict the shape of the acousticwavefront transmitted through the skull. Pre-focal field scans were performed to validate the backpropagation computational models between the focal plane and prefocal plane scans. Imaging distortion was charac- terized through a grid of single-point hydrophone measurements of synthetic aperture (SA) and single-transmit-focus(STF) imaging for the rat skulls with the 40-mm DMUA (operating bandwidth of 2.2 - 4.6 MHz). For all experiments, a range of frequencies relative to the human and ratmodel were chosen to accomplish a thorough broad- band characterization for all relevant frequencies (Human: 1.00,1.35,1.70,2.10 MHz; Rat: 1.80, 2.50,3.20,3.70,4.20MHz). RESULTS The hydrophone scans and backpropagation provide a thor- ough depiction of the transmission of focused ultrasound from the in- terior of the skull to the focalplane. The focal plane shows the Fig. 2 (abstract P35). See text for description distortion and loss of power caused by the skull. Figure 2 shows Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 75 of 122 the set point. All animals were survived for 3 - 5 days after tFUS applica- tion and observed for any abnormal behavior or adverse reaction. Histological evaluation was performed to determine whether the deliv- ered tFUS dose produced a BBB opening. For some animals, we ex- tracted the skull and performed in vitro transskull field mapping experiments to characterize the actual distortion to the tFUS beam in different planes with respect to bregma. RESULTS All animals that underwent the subtherapeutic tFUS applica- tions described above (over 30 Sprague Dawley rats 275 - 475 gm, male and female) have survived the procedure and no adverse events were recorded. Figure 2 shows an example of the rendering of the 3D im- aging data using the DMUA render engine. Onecan see the C-mode view with the lambda and bregma suture lines clearly visible with a thick arrow pointing to the bregma. The lines show the selected planes for tFUSapplication in a typical experiment. Figure 3 demonstrate the Fig. 3 (abstract P35). See text for description feasibility to visualize the tFUS beam access by rendering the DMUA el- ements and the skull (to scale) by using the results from the 3D scan. This allows for pre-treatment planning and post-treatment evaluation by bringing computational modeling and DMUA imaging data together in one computational model. Figure 4 shows an example result from a P36 spatiotemporal control of tFUS-induced temperature rise of 4°C for 10 Real-time spatiotemporal control of transcranial focused sec using the setup in Fig. 1. The pseudocolor overlay shows the spatial ultrasound in vivo distribution of the heated region and the spatiotemporal map (axial- 2,1 1 Dalong Liu , Emad S. Ebbini temporal) shows the localization in the axial direction. Electrical and Computer Engineering, University of Minnesota, CONCLUSIONS The results shown demonstrate the feasibility of real- Minneapolis, Minnesota, USA; Siemens, Seattle, Washington, USA time precise spatiotemporal control of tFUS dose application in a Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P36 temperature control application (e.g. drug-delivery BBB opening), but the DMUA could be easily used to control nonthermal application of Correspondence: Dalong Liu tFUS (e.g. neuromodulation). The results also demonstrate the unique Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P36 advantages of the DMUA approach, where DMUA imaging data pro- vides 3D volumetric rendering of the skull for target localization, pre- OBJECTIVES Transcranial focused ultrasound (tFUS) has been receiv- treatment planning and post-treatment evaluation. We envision a ing increasing attention by numerous research groups worldwide. It prescription-tFUS application using DMUA technology. is being investigated as a potential noninvasive treatment modality for tumor ablation, drug delivery through blood-brain barrier open- ing, Parkinson disease, etc. This renewed interest in tFUS canbe cred- ited to advances in diagnostic imaging and image-guidance modalities, especially MRI. The distortion of tFUS beam through the skull continues to be a major challenge to the ultimate goal of reli- ably localizing and controlling the therapeutic tFUS dose to meet the demands of precision therapy of neurological disorders. The goal of this study is to establish the feasibility of precise real-time spatiotem- poral control of tFUS energy application in a rat model in vivo. METHODS The 3.5-MHz dual-mode ultrasound array (DMUA) prototype was used to deliver tFUS to anesthetized animals under IACUC-ap- proved protocol. In each experiment, the animal was positioned prone on a stereotaxic unit with the head shaved and hair removed using de- pilatory cream to allow for good coupling with theDMUA through its water bolus (Fig. 1). A 3D scan of the skull was performed by mechanic- ally translating the DMUA prototype using a 3-axis motor (caudal-to- Fig. 1 (abstract P36). The dualmode ultrasound array (DMUA) used rostral). This scan was performed to identify the lambda and bregma in imageguided spatiotemporal control of tFUS in rat model in vivo suture lines, which were used a a reference for the treatment planes. The DMUA system provided a 3Drender interface to allow the visualization of these markers upon the completion of the 3D scan and before starting the tFUS application in selected planes (based on the- target circuitry for a given application.) All treatment planes were marked with respect to the bregma (e.g. bregma-2mm). Once the DMUA imaging/treatment slice was aligned with the desired treatment plane, imaging was performed before, during and after the application of the tFUS dose. Our system allowed for a full range of thermal and nonthermal control of tFUS by controlling the duty cycle. Spatial con- trol of the tFUS beam was provided by precision refocusing using a multichannel arbitrary waveform generation driver. In addition, ampli- tude modulation of tFUS on a frame-by frame basis (up to 500 fps) was achieved using a closed-loop control system basedon DMUA feedback. For the purposes of this paper, we describe the spatiotemporal control of subtherapeutic tFUS beams for thermal therapy. A typical tFUS shot wasbetween 4 - 15 seconds with initial exposure of approximately 400 W/cm2 in situ, designed to reach the temperature set point in ~0.5 sec. Fig. 2 (abstract P36). The 3D render interface used for visualization Ultrasound thermography basedon DMUA beamformed echo data of the bregma suture line (thick arrows) as reference for the from the target region was used for feedback. A PID controller was treatment planes (lines) employed to adjust the tFUS intensity to maintain the temperature at Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 76 of 122 Currently, there is no cure for PD patients and only a few options are available to alleviate symptoms. With advances in gene therapy re- search, adeno-associated virus (AAV) has the potential to serve as carrier to introduce therapeutic genes to the human body. In con- trast to invasive direct brain infusion, focused ultrasound (FUS) in combination with microbubbles has been shown to non-invasively and transiently open the blood-brain barrier (BBB). Here, our goal is to evaluate the potential neuro-protective and neuro-restorative ef- fects of non-invasive AAV-GDNF delivery in a MPTP mouse model. METHODS The PD mouse model was generated via intraperitoneal injec- tions of MPTP toxin at 30 μg/kg over five consecutive days. Animals were then divided into four groups (n = 7-10 per group): control, FUS only, AAV injection only, and FUS+/AAV+. For the FUS only and FUS+/AAV+ groups, both striatum and substantia nigra were sonicated unilaterally using a single element FUS transducer. For the AAV+/FUS+ group, a 100 μl mixture of AAV-GDNF vectors and polydispersed microbubbles were administered intravenously before sonication. Mice were allowed to sur- vive up to three months’ post sonication, which was followed by transcar- dial perfusion and tissue analysis. RESULTS Systemically administrated AAV vectors were capable of crossing the opened BBB and viral transduction was observed to be concentrated at the FUS targeted brain regions (i.e. striatum and sub- stantia nigra). The number of dopaminergic neurons at the point of sacrifice for each mouse was quantified by staining for tyrosine hy- droxylase (TH) in the substantia nigra regions. As shown in Fig. 1, mice that received a combination of AAV-GDNF and FUS exhibited Fig. 3 (abstract P36). A 3D isometric view of the DMUA elements significantly higher protection to the subsequent MPTP insult. In and the imaging/treatment slice with respect to the skull (as addition, neurorestoration was observed in AAV+FUS treated mice rendered from DMUA 3D imaging data) that was previously dosed with MPTP toxin. The dopaminergic neuron projections on the FUS+/AAV+ hemisphere also had higher density than the contralateral side. Behavioral study was performed 12 weeks after the initial unilateral treatment, where amphetamine- elicited unilateral rotation was observed in mice from the combined (AAV+FUS) treatment group (p < 0.05). CONCLUSIONS The findings of this study indicate the potential of gene delivery vectors for protecting and restoring the functions of dopaminergic neurons in PD and FUS as a non-invasive methodology for transcranial AAV delivery. Fig. 1 (abstract P37). a) Immunofluorescence TH staining revealed much more dopaminergic projections on the AAV+/FUS+ side of Fig. 4 (abstract P36). Typical result of realtime spatiotemporal the brain. b) Significantly higher number of TH+ neurons were found control of subtherapeutic tFUS thermal application in vivo rat model on the AAV+/FUS+ side of the brain compared to the contralateral side. c) Significantly higher dendrite density was observed on the AAV +/FUS+ side of the brain. d) Amphetamine-elicited behavioral studies revealed more frequent clockwise (toward the remaining lesion side) P37 rotation, signifying more prominent dopaminergic function on the Focused ultrasound-facilitated AAV-GDNF delivery for neuro-protection hemisphere receiving AAV+/FUS+ treatment and neuro-restoration in Parkinson ’s disease mice 1 1 1 Shutao Wang , Oluyemi Olumolade , Tara Kugelman , 2 1 2 Vernice Jackson-Lewis , Maria Eleni Karakatsani , Serge Przedborski , 1,3 P38 and Elisa E. Konofagou Transcranial focal passive detection and imaging: implementation Department of Biomedical Engineering, Columbia University, New York, on a multiple channel transmit/receive ultrasound phased array NY, USA; Department of Neurology, Columbia University, New York, NY, system USA; Department of Radiology, Columbia University, New York, NY, USA Chih-Hung Tsai, Hsin-Yu Chang, Chung-Han Wang, Hao-Li Liu Correspondence: Shutao Wang Department of Electrical Engineering, Chang Gung University, Taoyuan Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P37 City, Taiwan Correspondence: Chih-Hung Tsai OBJECTIVES Parkinson’s disease (PD) is the second most common Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P38 neurodegenerative disorder affecting millions of patients worldwide. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 77 of 122 OBJECTIVES Burst-mode focused ultrasound (FUS) exposure combined OBJECTIVES Low intensity focused ultrasound (LIFU) has the dem- with microbubbles (MBs) has been shown to induce temporal and local onstrated ability of non-invasively stimulating neural activity. This blood-brain barrier (BBB) opening. Contrast-enhanced imaging is now is of high value for therapeutic (stimulation, neuroprosthetics, served an indicator to postoperatively confirm the occurrence of BBB etc.) and diagnostic (preoperative mapping, etc.) purposes. A mul- opening. Developing a transmit/receive dualmode FUS apparatus has the tiscale simulation platform for image-based and personalized potential to observe focal position and fulfill implementation of real-time modeling of transcranial LIFU stimulation should be developed to monitoring of the occurrence of BBB opening. This study aims to disclose allow mechanistic studies, hypothesis formulation and testing, de- our recent development in using a self-designed multiple-channel trans- vice development, and, ultimately, personalized treatment plan- mit/receive system allow to perform passive cavitation analysis as well as ning, safety, and efficacy assessment. Experimental validation is to reconstruct focal beam distribution via passive imaging reconstruction. crucial to establish confidence in and explore the limitations of METHODS Homemade 256-channel ultrasound phased array driving sys- the modeling. tem was employed to drive a 256-channel FUS transducer to deliver focal METHODS The Sim4Life computational life sciences has been ex- transmit energy (fundamental frequency = 500 KHz, diameter = 120 mm, tended to: 1) Support full-wave acoustic simulation of transcranial curvature = 100 mm). The transmit pulse was designed to be 0.006 ms of sonication: For that purpose, new functionality to consider CT image- burst length, 2 Hz of pulse repeated frequency (PRF) and 3.56 MPa nega- based skull inhomogeneity information (density, speed-of-sound, and tive pressure output. Received circuits with the channel number ranging attenuation maps) and partly compensate for related focus aberra- from 16-64 were employed to perform RF signal receiving. During the in- tion and defocusing with multi-element transducer steering vitro experiments, either a strong needle reflector or microbubble (MB) optimization has been implemented. 2) Allow for neuronal dynamics tube was positioned with the flowed MBs concentrations been con- modeling: The NEURON library for compartmental neuronal dynamics trolled, the multiple channel RF signals were received in parallel with the modeling supporting detailed neuromorphology and channel dy- human skull were inserted. Passive cavitation detection was imple- namics has been integrated and parallelized simulations featuring mented, and passive imaging was reconstructed with the developed large numbers of neurons and neural networks can now be per- phase-corrected passive beamformed algorithm. formed. 3) Generate personalized, functionalized head models: Pa- RESULTS We demonstrate the feasibility in using this self-designed mul- tient image data can be used to generate anatomical geometries by tiple-channel system to serve as a platform to be operated at dual trans- segmentation and/or morphing of presegmented models. CT image mit/receive mode (Fig. 1). Multiple channels of RF data can be received in data informs on inhomogeneity, DTI image-data can be used for fiber parallel to reconstruct the passive imaging. The system now can support tracking to generate neuronal axon models, and Python tools facili- up to 64-parallel channel receiving for the following signal analysis and tate the anatomo-physiologically correct placement of cortical pyr- passive imaging formation. Point-spread function (PSF) imaging can be re- amidal neuron and deep brain stimulation relevant neurons (STN, constructed singly using 16-channel receiving, whereas higher channel GPi, IC). 4) Coupled acoustic and neuronal dynamics modeling: The provide superior SNR of imaging. The system also demonstrates the cap- Plaksin-Shoham-Kimmel (PSK) model of membrane-cavitation in- ability of the focal passive cavitation detection to real-time trace cavitation duced neurostimulation has been implemented and adapted for future activity specifically originating from the focal point. We also demonstrated use in combination with the compartmental cell models. Furthermore, that the implementation of a filtered phase-correction processing been coupled electromagnetic neuronal modeling is also supported. An ex- applied into the PSF reconstruction algorithm can successfully identify the perimental setup involving an acoustic transducer sonicating through a focal ultrasound deposition when penetrating through the skull. rat skull has been constructed. MicroCT image data has been acquired CONCLUSIONS We demonstrated the feasibility of the capability in and the 3D pressure distribution inside the skull has been measured using a self-built multiple-channel ultrasound transmit/receive system using computer controlled hydrophone scanning. to perform passive imagingand real-time focal PCD. The system and RESULTS The acoustic solver has been extensively validated previ- architecture has the potential to be developed to real-time monitor the ously against numerical and experimental data in homogeneous process of microbubble-facilitated FUS BBB opening process. setups and setups with homogeneous obstacles. The new experi- mental data allows for the first time successful validation of a setup involving inhomogeneous media using the previously pre- sented Gamma method for uncertainty assessment-based, object- ive comparison of 3D pressure distributions (Fig. 1).The neuron functionalized anatomical head models have been partly vali- dated by comparing modeling of transcranial electric/magnetic and deep brain stimulation with experimental data from literatur- e.The PSK-model could be simplified without significant impact on the results, thus enabling its integration into 3D, extended, morphological cell models. For that purpose, the previously bidir- ectional coupling of the electrophysiological and cavitation me- chanics parts has been broken up and further separation allows to pre-compute costly parts of the model, accelerating the mod- eling by more than one order of magnitude. Fig. 1 (abstract P38). Passive imaging reconstruction: upper row CONCLUSIONS A multi-scale framework for the computational inves- shows the tube was filled with air, and lower row shows tube was tigation of LIFU neuro-stimulation is being developed. It features filled with MBs image-based acoustic propagation modeling (including support for bone inhomogeneity) and focusing functionality, patient-specific functionalized anatomical model generation with realistic neuron P39 placement, integrated neuronal dynamics simulation, and a Multiscale Modeling of transcranial focused ultrasound coupled model of LIFU-induced neurostimulation that is currently neurostimulation and experimental validation: initial results still limited to 0D neuron models, but has been prepared for the fu- 1,2 2 1 3 Hazael Montanaro , Mehmet S. Özdas , Esra Neufeld , Théo Lemaire , ture modeling of 3D-extended, morphologically-detailed physio- 3 2 1, 2 Silvestro Micera , Mehmet F. Yanik , Niels Kuster logical neuron models. Important parts of the platform (acoustic Computational Life Sciences, IT'IS Foundation for Research on and neuronal simulations, functionalized models) could be success- Information Technologies in Society, Zurich, Zurich, Switzerland; Swiss fully validated against new and existing experimental data.The pre- Federal Institute ofTechnology (ETHZ), Zurich, Switzerland; Swiss Federal sented progress is an important step towards the goal of allowing Institute of Technology (EPFL), Lausanne, Switzerland mechanistic studies, hypothesis formulation and testing, device de- Correspondence: Hazael Montanaro velopment, and, ultimately, personalized treatment planning, safety, Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P39 and efficacy assessment. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 78 of 122 ofPSPIO and pDNA onto MBs were 134.5 g and 17.1 g, individually. Figure 1B shows that PSp-MBs with ultrasound could achieve gene transfection and the expression of gene could be further enhanced by MT process. We also confirmed that the PSp-MBs could perform successful BBB opening without bioeffects by the trigger ofultra- sound with an acoustic pressure of 0.3 MPa (Fig. 1C). CONCLUSIONS The MBs have fairly payloads of PSPIO-pDNA. We demonstrated that PSp-MBs with ultrasound can perform locally gene delivery and open BBB concurrently. In addition, the efficiency of gene delivery could be further enhanced by MT process. Future works include quantifying and tracking of distribution of gene deliv- ery and Parkinson’s disease rats via magnetic resonance imaging. Fig. 1 (abstract P39). Cortex subregion of a human model with anatomically positioned pyramidal neurons and visualized transmembrane voltage activity P40 SPIO-PEI-pDNA complex loaded microbubbles for ultrasound-based gene therapy in brain Chun-Yao Wu, Ching-Hsiang Fan, Rih-Yang Huang, Chien-Wen Chang, Chih-Kuang Yeh Department of Biomedical Engineering and Environmental Sciences, National Tsing Hua University, Hsinchu, Taiwan Correspondence: Chun-Yao Wu Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P40 OBJECTIVES Recently, gene therapy has attracted much attention espe- cially in neurodegenerative diseases. Currently, gene delivery within cen- tral nerves system mainly relies on invasive intracerebral injection or viral vectors to circumvent the obstacle of blood-brain barrier. Non-viral gene delivery via systematic transvascularroute is an attractive alterna- tive since it is non-invasive. However, a high-yield and targeted gene delivery platform is still lacking. In order to improve the efficiency of gene delivery, this study proposed an ultrasonic sensing vector for gene delivery into brain through polyethylenimine (PEI)-superparamagnetic Fig. 1 (abstract P40). (A) Microscope images of the PSpMBs with iron oxide (SPIO)-pDNAloaded microbubbles (PSp-MBs). Cooperating Prussian Blue and propidium iodide staining. The colocalization of with ultrasound exposure, PSp-MBs could transport the PSp nanoparti- the PSp nanoparticles in the Prussian Blue and propidium iodide cles into the desired brain region by acoustic MBs cavitation activity. The staining indicated a good conjunction of PSPIOpDNA and MBs. (B) rate of gene transfection would be enhanced by the modification of PEI Gene transfection efficency of PSpMBs with different conditions, onto PSp nanoparticles. In addition, by an externally applied magnetic suggesting that PSpMBs with ultrasound could achieve gene field, magnetic targeting (MT) can further increase the deposition of PSp transfection. (C) BBB disruption of PSpMBs cooperating with FUS by at the targeted location, enhancing the gene delivery. different parameters METHODS The PSPIO was consisted of PEI molecular and SPIO nano- particles (diameter: 10 nm) via ligand exchange. The PSPIO were then conjugated with pDNA and loaded onto the lipid surface of MBs by electrostatic force. PSPIO-pDNA (luciferase plasmid) modulated onto the MBs was confirmed by Prussian blue staining and propidium iod- P41 ide staining. The size, concentration and PSPIO payload were mea- Numerical study of bubble area evolution during acoustic droplet sured by multisizer and plate reader, respectively. C6 glioma cell and vaporization enhanced HIFU treatment 1 1 1 2 Sprague-Dawleyrats (N = 4) were used in this study. The gene trans- Ying Xin , Aili Zhang , Lisa X. Xu , Jeffrey B. Fowlkes fection efficiency and BBB opening region resulted from PSp-MBs School of Biomedical Engineering, Shanghai Jiao Tong University, with ultrasound (frequency = 1 MHz, energy = 0.1-0.5MPa, cycle = Shanghai, China; Department of Radiology, University of Michigan 5000, PRF = 1 Hz, sonication time = 60 s) were evaluated by bio- Health System, Ann Arbor, Michigan, USA luminescence imaging and Evans blue staining, individually. The MT Correspondence: Ying Xin process was performed by a 0.48 Tesla external magnet. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P41 RESULTS Figure 1A shows the fabricated PSp-MBs. The co- localization of the PSp nanoparticles in the bright field images and OBJECTIVES Acoustic Droplet Vaporization (ADV) has the potential to the fluorescent image indicated a good conjunction of PSPIO-pDNA shorten treatment time of high intensity focused ultrasound (HIFU) and MBs (as arrows). The mean size and concentration of PSp-MBs while minimizing thepossible effects of microbubbles along the were 1.5 m and (5-10) × 109 bubbles/mL, respectively. The payload propagation path. Distribution of the bubbles formed from the Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 79 of 122 droplets during the treatment is the major factor shaping the therapeutic region. However, there is no simulation report of this phase shift droplets assisted HIFU therapy, in which the bubbles form dynamically and only exists in places where certain acoustic conditions are met. In order to provide an approach for compre- hensive parametric analysis, which could save time and effort in future studies, simulation of the dynamic formation of the bub- bles inside the tissue during this treatment is carried out in this paper. The proposed model is verified using previously published experimental results. Numerical results predicting the effect of the presence of a preformed bubble wall are also obtained and discussed. METHODS The schematic of the model setup and the axisymmetric geometry of the simulation spatial domain are shown in Fig. 1. The Fig. 1 (abstract P41). (a) Schematic of the physical model and (b) origin of coordinate system coincides with the HIFU geometric focus. the geometry of the axisymmetric simulation spatial domain The phase shift droplets are assumed to be distributed uniformly in the phantom. The vaporized bubbles also distribute evenly, which is expected in phantoms and likely the case for larger bubbles trapped within the vasculature in tissue. The average diameters of droplets are usually smaller than 5 microns (Kripfgans et al., 2000; Zhangand Porter, 2010), and the formed bubbles have a diameter 5 times of the one of the droplets based on ideal gas law, which is much smaller than the ultrasound wavelength (about 1.9 mm to 0.5 mm when the frequency is in the therapeutic range of 750 kHz to 3 MHz), so the bubble area is treated as a homogeneous medium. Linear sound equation with attenuation in the frequency domain is used to describe the acoustic pressure p. The density ρ, sound speed c and attenuation coefficient α are functions of the gas void fraction fG, which can be calculated directly from the bubble size distribution (Church, 1995). ADV droplets would be vaporized under certain con- ditions, which are related to ultrasound frequency, diameter of the droplets, super heat degree and total ultrasound ‘ontime’. The follow- ing assumptions are used to decide how much bubbles will form in certain acoustic field:(i) Threshold for droplets in phantom gel ex- posed to 750 kHz ultrasound is found to be 7.6 MPa as was found in experiment.(ii) The probability of droplets vaporizing increases linearly with rarefactional pressure in a range, which is function of the threshold according to the experimental results (Lo, et al, 2007). (iii) The incident wave got reflected at the interface of the bubble Fig. 2 (abstract P41). (a) Droplets size distribution used in the area based on the difference of the acoustic impedance of the two simulation. (b) Sound speed and (c) sound attenuation coefficient medium. Meanwhile the bubble area presents different impedance as a function of void fraction in the phantom. (d)Sound speed as to different frequency. So the nonlinear part of the acoustic field is function of frequency (void fraction is 2.52×104 mL/mL) considered and simplified based on it. RESULTS The ultrasound used in the model is 750 kHz, with a focal pressure of 9.8 MPa and 14.7 MPa. Each pulse is 20 μs. The droplets concentration is 1.3×106/mLand the size distribution of the droplets is shown in Fig. 2(a). The calculated sound speed and sound atten- cuation is shown in Fig. 2(b-d) The sectional view of pressure field and bubble area after 20 cycles of ultrasound exposure when focal pressure is 9.8 MPa is shown in Fig. 3. The sectional view of pressure field and bubble area after 200 cycles of ultrasound exposure when focal pressure is 14.7 MPa is shown in Fig. 4 (a, b). All the pressure was representing in dB scale using 9.8 MPa as the reference pressure. These results are close to the experimental result (Lo. at al, 2006) that they can describe the unique feature of the bubble area when the in- cident pressure varies. A layer of bubble has been proved to provide protection of the distal area in HIFU therapy. This model can also predict the bubble area when a bubble layer is created before treat- ment in the therapy (Fig. 4 (c, d)). CONCLUSIONS This model can describe the bubble area evolution during the acoustic droplet vaporization enhanced HIFU therapy Fig. 3 (abstract P41). The sectional view of (a) pressure field and (b) while using a simple linear acoustic model. The model can also be bubble area after 20 cycles of 20 μs pulses of ultrasound exposure, used in the case when a bubble layer pre-exists to protect the distal of which focal pressure is 9.8 MPa and frequency is 750 kHz. The area from the HIFU ablation. The heating effect can be coupled to pressure was represented in dB scale using 9.8 MPa as the reference the model in the future and then it can be used as a planning tool in pressure. The white scale bar represents 2 mm ADV enhanced HIFU therapy and predict the treatment outcome. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 80 of 122 different FUS pulses ([peak-negative pressure (Pneg): 8, 12MPa, pulse length (PL): continuous wave], [Pneg: 25MPa, PL: 5 ms, PRF: 20 Hz]) wereemitted. Dye movement (i.e., clearance) at the hydrogel surface was observed with a video camera. METHODS Our initial challenge was to create an appropriate material for studying acoustic streaming. We hypothesised that existing hydrogels (e.g., polyacrylamide and gelatin) used as ultrasound phan- toms, mimic the acoustic properties of tissue, but not the tissue microenvironment. We created three phantoms – gelatin, polyacryl- amide (PAA), and a new macroporous polyacrylamide (MPPA) hydro- gel – and analysed their structure using scanning electron microscopy (SEM) and mercury intrusion porosimetry (MIP). Based on our findings, we selected the MPPA for our tissue-mimicking material. Focused ultrasound (FUS) pulses were emitted from a single-element transducer [centre frequency (fc): 5 MHz], which was driven by a function generator through a power amplifier. The focal point of a FUS transducer’s beam (axial FWHM: 3.2 mm, lateral FWHM: 0.45 mm) was placed at the distal surface of the MPPA. A model drug (Bromophenol blue) was injected in and around the focal region and different FUS pulses ([peak-negative pressure (Pneg): 8, 12MPa, pulse length (PL): continuous wave], [Pneg: 25MPa, PL: 5 ms, PRF: 20 Hz]) wereemitted. Dye movement (i.e., clearance) at the hydrogel surface was observed with a video camera. RESULTS SEM revealed that although gelatin and PAA were porous, they Fig. 4 (abstract P41). The sectional view of (a) pressure field and (b) had pockets of water separated by walls of material (Fig. 1a, b). This sug- bubble area after 200 cycles of 20 μs pulses of ultrasound exposure, gests that acoustic streaming with these materials are unlikely to occur of which focal pressure is 14.7 MPa andfrequency is 750 kHz. The unless enough stress is applied to break the walls. In contrast, MPPA had sectional view of (c) pressure field and (d) bubble area when a 1 interconnected pores throughout the gel – a feature that more accurately mm thick of bubble layer pre-existed. The pressure was represented resembled the interstitial space of soft tissue (Fig. 1c). The gelatin, PAA in dB scaleusing 9.8 MPa as the reference pressure. The white scale and MPPA had a porosity of 81, 76 and 88%, respectively; a permeability bar represents 2 mm of 2400, 1500 and 6500 mdarcy, respectively. We then selected MPPA for further study, because it most accurately mimicked the tissue microenvir- onment. Using focussed ultrasound in continuous wave (Pneg: 12MPa), P42 we directly observed acoustic streaming in the MPPA material in the form Noninvasive and localised acoustic micropumping – an in vitro of the dye progressively clearing from the focal region (Fig. 2). The clear- study of an ultrasound method that enhances drug distribution ance rate is represented by the change in optical density over time (Fig. through a physiologically-relevant material 1 1 2 3).Pulsedultrasoundwasapplied(Pneg:25MPa,PL:5ms,PRF:20 Hz) Ahmed Elghamrawy , Florentina de Comtes , Hasan Koruk , 3 1 having the same average intensity as 8MPa continuous wave and an in- Ali Mohammed , James J. Choi crease in clearance rate was observed. Moreover, pulsed ultrasound re- Department of Bioengineering, Imperial College London, London, UK; 2 3 ducedunwantedthermaleffects. Mechanical Engineering, MEF University, Istanbul, Turkey; Materials, CONCLUSIONS Our results reveal the first direct observations of acous- Imperial College London, London, UK tic streaming in a soft tissue microenvironment. We showed that al- Correspondence: Ahmed Elghamrawy though gelatin and polyacrylamide have acoustic properties suitable Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P42 for ultrasound imaging and hyperthermia experiments, they are not ef- fective in studying acoustic streaming. Instead, we recommend the use OBJECTIVES Acoustic streaming - the displacement of fluid by sound of MPPA, because it supports fluid flow through its microstructure by - has been proposed as the mechanism for therapeutic effects, such having interconnected pores or channels. Using this phantom material, as drug distribution enhancement and neurostimulation, yet there we wereable to study acoustic streaming under different exposure has been no direct observation or characterisation of this effect in parameters; finally demonstrating that pulsed ultrasound pro- soft tissue microenvironments, making it difficult to optimise and duces greater acoustic streaming with less risk ofthermal damage control. Post-mortem and indirect analyses have revealed changes in than with continuous wave emission using the same acoustic en- dye distribution and neuronal excitation, but it has remained uncer- ergies. Our future work is to study optimise ultrasound parame- tain whether these biological outcomes were due to acoustic stream- ters to maximise acousticstreaming in this material and in in-vivo ing or other responses, such as acoustic cavitation. We aimed to be tissue for enhancing drug distribution and neurostimulation. the first to directly observe ultrasound-induced streaming during sonication in a tissue-mimicking material. METHODS Our initial challenge was to create an appropriate material for studying acoustic streaming. We hypothesised that existing hydrogels (e.g., polyacrylamideand gelatin) used as ultrasound phan- toms, mimic the acoustic properties of tissue, but not the tissue microenvironment. We created three phantoms – gelatin, polyacryl- amide (PAA), and a new macroporous polyacrylamide (MPPA) hydrogel – and analysed their structure using scanning electron mi- croscopy (SEM) and mercury intrusion porosimetry (MIP). Based on our findings, we selected the MPPA for our tissue-mimicking material. Fig. 1 (abstract P42). Scanning electron microscopy of (a) gelatin, Focused ultrasound (FUS) pulses were emitted from asingle-element (b) polyacrylamide and (c) macroporous polyacrylamide. Zoomed in transducer [centre frequency (fc): 5 MHz], which was driven by a sections show closed pockets or pores for gelatin and PAA but, function generator through a power amplifier. The focal point of a MPAA pore structure provides fluid a path to follow throughout the FUS transducer’s beam (axial FWHM: 3.2 mm, lateral FWHM: 0.45 gel. Fluid can travel from the pore on the gel surface to the deeper mm) was placed at the distal surface of the MPPA. A model drug pore behind (bottom right rectangular areas) (Bromophenol blue) was injected in and around the focal region and Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 81 of 122 and synthesised multifunctional human serum albumin-chlorin e6 nanoassemblies (HSA-Ce6 NAs) that integrate imaging andtherapy functionalities into a single nano-platform for MRI guided sonody- namic therapy. We show the albumin-sonosensitizer nanoassemblies is highly accumulated in tumor tissue and had highly effective in destructing cancer cells under ultrasound and MRI-visible in vivo and thus, suitable for theranostic applications in cancer. METHODS HSA-Ce6 NAs synthesis as follows: after dissolution of HSA Fig. 2 (abstract P42). From images acquired, differences images at in water, Ce6 was added into the solution to precipitate the HAS- 1, 5 and 10 s after the start of the sonication (Pneg: 8MPa, PL: cotinuous Ce6, Then HSA-Ce6 NAs were mixed with MnCl2 solution at a molar wave) were obtained (a-c). The black circle denotes the sonicated ratio of 2:1 for the HSA-Ce6 NAs.To detection of ROS in HSA-Ce6 region, as the darker dye is cleared from hydrogel, the lighter clearer gel NAs, different samples were mixed with 1 mM 2′,7′-dichlorofluorescin surface is shown causing an increase in optical density diacetate (DCFH-DA), and then irradiated by different US times or US intensities.In vitro SDT: The ultrasound transducer (diameter: 1.5 cm, resonance frequency: 0.5 MHz, duty factor: 50 %, interval: 100 ms; ultrasonic intensity: 0.5 W/cm2) was fixed at the bottom of water bath. The cell culture plate was suspended 10 cm above the ultrasound transducer. In vivo MRI: The in vivo MRI imaging experiments were acquired using a 3.0 T clinical MR scanner (TIM TRIO, Siemens, Germany) with a small animal coil. T1-weighted MR images were acquired using the following parameters: TSE sequence, TR= 700 ms, TE =13 ms, FOV =32 × 45 mm, slice thickness= 1 mm and flip angle=180°. RESULTS The sonodynamic effect of HSA-Ce6 NAs was confirmed by measuring reactive oxygen species using DCFH-DA as a detector, as shown in Fig. 1, thefluorescence intensity of ROS exhibits a time- dependent and an intensity-dependent enhancement, indicating ROS from HSA-Ce6 NAs upon ultrasound irradiation. Thecell uptake behavior of HSA-Ce6 NAs was investigated through confocal micros- copy, the HSA-Ce6 NAs has high cell uptake efficiency with high fluorescence intensityin U87 cells (Fig. 2). Moreover, the SDT in vitro of HSA-Ce6 NAs could be further improved. It was found that single HSA-Ce6 NAs or ultrasound treatment could only induce partial cell death at the current conditions. In marked contrast, the combination treatments (SDT) were found to be highly effective in destructing cancer cells. Indicating that ultrasound has a good penetrability and SDT effect (Fig. 3). In addition, the HSA-Ce6 NAs can serve as chelat- Fig. 3 (abstract P42). Clearance rate is the average of optical ing agents to capture Mn2+ for MRI imaging by forming stable che- density values of the pixels in the black circle (Fig. 2) over time. As lates. The in vivo T1-weighted MR imaging of U87 tumor-bearing continuous wave ultrasound pressure increases from Pneg: 8MPa mice was carried out after iv injection of HSA-Ce6 NAs, The T1 signals (red) to 12MPa (green) there is a slight increase in dye clearance of tumor on mice strengthened with the increase of time interval, rate. Pulsed ultrasound was applied (Pneg: 25MPa, PL: 5 ms, PRF: 20 and reached a peak after 24 h postinjection of HSA-Ce6 NAs (Fig. 4). Hz) having thesame average intensity as 8MPa continuous wave This indicated the HSA-Ce6 NAs could provide the optimal time win- and an increase in clearance rate was observed (blue). All applied dow for sonodynamic therapy so that the ultrasound irradiation parameters initiated dye clearance compared to whenno ultrasound could be conducted in the targeted lesion. was applied (black) CONCLUSIONS We have shown our human serum albumin-chlorin e6 nanoassemblies are indeed visible by MRI in vivo and that they can be targeted by FUS to deliver and release reactive oxygen spe- cies (ROS) to kill cancer cells, paving the way for their theranostic ap- P43 plications under MRI-guided sonodynamic therapy. MRI-guided focused ultrasound mediated smart albumin- sonosensitizer nanoassemblies for sonodynamic therapy of glioma 1 1 2 1 2 Qian Wan , Dehong Hu , Mengjie Chen , Zonghai Sheng , Jun Zhou , 1 1 Xin Liu , Hairong Zheng Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences, Shenzhen, Guangdong, China; Medical College of Chinese Three Gorges University, Yichang, China Correspondence: Qian Wan Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P43 OBJECTIVES Sonodynamic therapy (SDT) represents an emerging ap- proach that offers the possibility of non-invasively eradicating solid tumors in a site-directed manner, which involves the synergistic ef- Fig. 1 (abstract P43). The ROS generation of HSA-Ce6 NAs under fect on cell damage by the combination of the sonosensitizer and US irradiation. (A) Fluorescence emissin spectra of HSA-Ce6 NAs in ultrasound, The sonosensitizer is the key factor of SDT.Chlorin DCFH-DA solution with the increase of US irradiation time. (B) e6(Ce6) has been widely used as a sonosensitizer in sonodynamic Fluorescence emissin spectra of HSA-Ce6 NAs in DCFH-DA solution therapy (SDT) on many human tumors, indicating Ce6 possesses ex- with the increase of US irradiation intensity cellent sono-activities with tiny toxicity.In this study, we designed Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 82 of 122 bony tissues. To devise ascheme for performing thermometry based on these MR properties, we first must know the relationship between these variables and temperature. The first objective of this investiga- tion was to quantify these relationships in ex-vivo tissue, with the eventual goal of developing a strategy for performing such measure- ments in-vivo during FUS sonication. To date, such measurements have been performed using 3D ultrashort echo time (UTE) acquisition pulse sequences, which severely limits the achievable temporal reso- Fig. 2 (abstract P43). Fluorescence images displayed cellular lution. Therefore, our second objective was to explore the feasibility localization of HSA-Ce6 NAs after 3 incubation with HSA-Ce6. Red of performing accurate temperature measurements in cortical bone represented the fluorescence HSA-Ce6NAs. Blue represented the using much faster 2D acquisitions with slice selective RF pulses. fluorescence of DAPI and green represented the fluorescence of the METHODS We measured the temperature dependence of T1 and cell membrane of U87 cells T2* in cortical bone from a recently euthanized swine tibia and bovine long bone using both 2D and 3D pulse sequences. In order to accurately measure these MR properties, we had to cre- ate a system in which we could vary the temperature of the bone as well as keep the temperature constant within the MR scanner during MR imaging. To achieve this, we built a system that consisted of an MR-safe water bath, a combination water- heater/recirculation pump that regulated the temperature of the water bath, and fiber-optic temperature sensors that monitored the temperature of a bone sample immersedin the water bath (Fig. 1). The bone sample was placed inside of the regulated water bath and its temperature was monitored by the fiber optic sensors while we acquired our measurements that would allow us to calculate the temperature dependence of the MR varia- bles.T1 was measured at 3T by acquiring a series of spoiled gra- dient-echo ultrashort UTE images, each having the same TR and Fig. 3 (abstract P43). The in vitro SDT in U87 cells. (A) Schematic TE but different flip angles, and then fittingthe measured signal- diagram of the insonation device. (B) Quantitative evaluation of cell versus-flip angle curve to the theoretical dependence at each survivals of each group voxel contained entirely within the bone. In order to find this theoretical dependence, we derived an equation that would give us the steady-state incoherent signal as a function of T1 and T2*. T2* was measured by acquiring a series of spoiled gradient-echo images, each having the same TR and flip angle but different echo times, and then fitting the measured signal-versus-TE curve to a mono-exponential decay at each voxel contained entirely within the cortical bone. The T1 and T2* measurements were per- formed at different temperatures ranging from below body temperature to nearly60°C. The temperature of the water bath was held constant during each T1 and T2* measurement. Fig. 4 (abstract P43). In vivo MRI images of the mice bearing U87 RESULTS T1 and T2* were both found to increase approximately linearly tumor after iv injection of HSACe6 NAs at different times. I.V. with temperature. The slope of this linear dependence was measured to injection dose of HSACe6 NAs is 2.0 mg Ce6/kg be 1.25 ms/°C for T1and 4.66 μs/°C for T2* in swine tibia and 1.38 ms/°C for T1 and 3.69 μs/°C for T2 in bovine long bone, utilizing the 3D imaging pulse sequence (Fig. 2). Additionally, we measured T1 and T2* in bovine bone using the 2D imaging pulse sequence, yielding slopes of 0.74 ms/°C P44 for T1 and 4.26 μs/°C for T2* (also Fig. 2). The quality of the theoretical fits Towards rapid MR thermometry in cortical bone 1 1 2 3 to the measured signal was generally excellent (Fig. 3). Phoebe Miller , Sina Tafti , David Keder , Quinton Miller , Darius 3 1 CONCLUSIONS The T1 variation with temperature was found to Hossainian , Wilson Miller be similar in swine and bovine cortical bone when measured Radiology and Medical Imaging, University of Virginia, Charlottesville, using a 3D pulse sequence with a nonselective excitation RF Virginia, USA; Physics, University of Virginia, Charlottesvle, Virginia, USA; pulse. However, the T1 values measured using a 2D pulse se- Biomedical Engineering, University of Virginia, Charlottesvle, Virginia, quence with a slice-selective excitation RF pulse were vastly dif- USA ferent. This discrepancy is almost certainly due to naive Correspondence: Phoebe Miller application of the theoretical signal versus T1 relationship, which Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P44 does not account for the non-ideal slice profile of our slice-select- ive excitation RF pulse. Such effects must be considered when OBJECTIVES Bone metastases are an important, FDA-approved treat- performing T1-based thermometry using 2D acquisitions. By con- ment target for MR-guided focused ultrasound therapy. However, it trast, the T2 variation with temperature was found to be similar is not currently possible to noninvasively measure temperature in when the same sample was measured using 2D and 3D acquisi- bony tissues. Conventional PRFS-based MR thermometry is likely not tions. Thus it may prove to be more straightforward to achieve feasible in cortical bone due to very fast T2* signal decay. Other MR quantitative accuracy when performing T2-based thermometry using properties, including T1 and T2*, also depend on temperature and 2D acquisitions. Whereas our 3D pulse sequence required 1 minute to thus offer alternative pathways for performing MR thermometry in Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 83 of 122 acquire a single image, our 2D pulse sequence required only 2.5 sec- OBJECTIVES In recent years, with the rapid development of medical onds. This time scale is much more favorable for monitoring imaging, ultrasound molecular imaging becomes one of the hot temperature changes during FUS application in vivo. spots in molecular imaging research field. METHODS The design of molecular probes is the key point and prerequisites for ultrasound molecular imaging. People increas- ingly pay more attention to the targeted ultrasound contrast agents which are the ultrasound molecular probes. And the inter- section of multiple disciplines will promote the development of ultrasound molecular imaging. RESULTS It is also urgent to develop a set of special equipment for efficient ultrasound molecular imaging. CONCLUSIONS The ultrasound molecular imaging instrument, micro- bubble/microsphere triggered device, imaging monitoring and ultra- sound molecular probes can be integrated into the low intensity ultrasound molecular imaging and therapy system, which will hope- fully bring about the integration of ultrasound molecular imaging, in vivo drug deliveryand controlled release and evaluation of treat- ment efficacy. It provides an innovative research platform for ultra- sound molecular imaging and therapy. Fig. 1 (abstract P44). Experimental apparatus P46 MR and fluorescence dual-modal imaging-guided sonodynamic therapy of gliomas through a multifunctional theranostic nanoplatform 1 2 1 Fei Yan , Meijun Zhou , Hairong Zheng Shenzhen Institutes of Advanced Technology, Shenzhen, China; Department of Ultrasonography, The Third Affiliated Hospital of Southern Medical University, Guangzhou, China Correspondence: Fei Yan Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P46 OBJECTIVES The aim of this study is to develop a multifunctional theragnostic nanoplatform for MR and fluorescence dual-modal im- aging-guided sonodynamic therapy of gliomas. METHODS Sinoporphyrin sodium (DVDMS), a NIR-absorbing sonosen- Fig. 2 (abstract P44). T1 and T2* dependence on temperature in sitizer and photosensitizer, was firstly used to chelate with manga- both swine and bovine bones when measured with 3D techniques, nese ion (Mn2+) and then encapsulated into liposomes by thin-film and in bovine bone when measured with 2D techniques rehydration method to fabricate DVDMS-Mn-Liposomes (DVDMS-Mn- LPs). The characterizations of DVDMS-Mn-LPs, their imaging capabil- ity in vitro and in vivo and SDT effect in subcutaneous and orthotopic glioma mouse models were examined (Fig. 1). RESULTS The resulting nanoparticles are proved to be physiologically stable and biocompatible, allowing time-dependent and intensity- dependent generation of oxygen free radicals upon ultrasound irradi- ation. Good T1-weighted MR and fluorescence imaging capabilities were demonstrated. We further employed this nanoparticle to treat subcutaneous and orthotopic glioma, demonstrating that SDT with DVDMS-Mn-LPs significantly improved the anticancer effect than that of PDT with DVDMS-Mn-LPsin the presence of skull. Histological ana- lysis further revealed much more apoptotic cells and lower tumor cell proliferation, confirming the advantageous anti-tumor effect of SDT over PDT against glioma. CONCLUSIONS Our study developed a novel theragnostic nanoplat- Fig. 3 (abstract P44). T1 and T2* measurement data, together with form and provided a promising strategy for imaging-guided SDT for best fit curves, for swine tibia glioma treatment. P45 Multi-disciplinary integration of ultrasound molecular imaging Zhigang Wang Institute of Ultrasound Imaging, Chongqing Medical University, Chongqing, China Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P45 Fig. 1 (abstract P46). See text for description Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 84 of 122 P47 OBJECTIVES Biofilm's control is a critical issue for water treatment Comparative study on the characteristics of PLGA microspheres systems. The primary goal is to use low-intensity pulsed ultrasound with different drugs to promote liposomes which encapsulate antibiotics into agarose- Dong Yu film based bio-film phantoms and release the encapsulated antibi- Chongqing Medical University, Chongqing, China otics by applying mild-intensity focused ultrasound. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P47 METHODS Liposomes were synthesized using 1,2-Diacyl-sn-Glycero-3- Phosphocholine (PC) and 1,2-Dipalmitoyl-sn-Glycero-3-Phosphoethanola- OBJECTIVE To compare the characteristics of PLGA microspheres mine (DPPE) mixedto a percent molar ratio of 95:5 and dissolved in containing different drugs and green fluorescent dextran, to provide chloroform in a round-bottomed flask and then were dialyzed against the basis for drug treatment of fungi. PBS for 24 hours. Dialysate (one liter) was changed 4-6 times, resulting in METHODS PLGA microspheres were prepared by double emulsification a relatively pure liposome suspension that encapsulate a gentamicin so- method, and different antifungal drugs were encapsulated, as well as green lution, Liposome size distributions were determined by the autocorrel- fluorescent dextran. The experiment consisted of PLGA microspheres, PLGA ation function of the dynamic light scattering method. The1.5% sodium microspheres containing green fluorescent dextran, PLGA microspheres alginate solution was prepared by adding 3g alginic acid, sodium salt loaded with amphotericin B and PLGA microspheres loaded with flucona- and 0.6mL glycerol to 200 ml distilled water in a 250ml beaker via con- zole. The surface morphology, internal structure, particle size, potential, en- tinuous vortex-mixing overnight. Biofilms were generated using 1x106 trapment efficiency and drug loading were measured and compared. Ralstonia insidiosa in 1.5% alginate solution and allowed to stand for 5 RESULTS (1) The PLGA microspheres were uniform in size, well dis- minutes to assure that the surface of the solution flat. 600 ml of 2% CaCl2 persed and spherical in shape and regular in morphology. (2) The was added to the alginate from the top for two hours. After particle size of PLGAmicrospheres was 343.17 ± 44.5nm. (3) PLGA polymerization, the film was washed with sterile water and incubated microspheres containing green fluorescent dextran were used as the with R2B overnight at room temperature. A two-step treatment was ap- experimental control group, and the green fluorescence was ob- plied. The first step is the penetration of the liposomes into the biofilm. served under laser confocal microscope. Indicating that PLGA micro- The front part of the transducer was immersed in the liposome suspen- spheres successfully encapsulated green fluorescent dextran. sion. The distance between the transducer surface and the top of the al- CONCLUSIONS PLGA microspheres, PLGA microspheres loaded with ginate-flm was 1cm.Tone-bursts ultrasound (2.25MHz with duty cycle PLGA and PLGA microspheres have good physical properties under 10%) were emitted downward into the liposome solution. Experiments the same conditions, which lays a foundation for the synergetic ef- were conducted with ultrasound spatially and temporally averaged inten- fect of microbubbles in the treatment of fungi. sity, ISATA = 0.14 W/cm2.The ultrasound transducer used is a single-elem- ent non-focusing piezo-ceramic transducer operated at 2.25 MHz, with active radius a =10 mm. An arbitrary waveform function generator was P48 programmed to produce a tone-burst sinusoidal signal of duty cycle of Therapeutic effect of focused ultrasound combined with anticancer 10% for 1 min insonation duration, and the output of the waveform gen- drug loaded microbubble complex for pancreatic cancer: erator was used as the input of a 55 dB RF power amplifier whose output preliminary study was used to drive the transducer. The experiment step two uses a fo- Eun-Joo Park, Yun Deok Ahn, Yuri Cheon, Jae Young Lee cused ultrasound to burst the liposomes inside the biofilm and release Radiology, Seoul National University Hospital, Seoul, Korea the drug from the liposomes in situ. The sample petri dish was mounted Correspondence: Eun-Joo Park and submerged in a tank containing filtered, distilled and degassed Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P48 water, the position of the sample petri dish can be transported in three orthogonal dimensions by a computer-controlled manipulator with step OBJECTIVES As focused ultrasound (FUS) combined with microbub- distance of 1 mm. A piezo-ceramic transducer operating at f0 =1.1 MHz bles has been widely studied in cancer treatment, there is growing (geometric focal length d = 62 mm, active radius, a = 32 mm) was interests in developing nanoparticles for FUS enhanced cancer drug mounted at the bottom of a tank filled with degassed water facing up- delivery. This study was designed to evaluate therapeutic effects of ward. The distance between the transducer and the biofilm wasadjusted anticancer drug loaded microbubble complex in combination with to be equal to the geometric focal length (d=62mm). The same signal focused ultrasound treatment for pancreatic cancer. generator and amplifier were used, ultrasound tone-bursts (1.1MHz with METHODS Immunodeficient mouse inoculated with CFPAC-1 were duty cycle 10%) were emitted upward to the biofilm. The sample was used as the pancreatic xenograft model for in vivo studies. Animals moved in a horizontal plane near the focal plane of the source transducer were treated in five groups: control, Doxorubicin-only (Dox), Doxo- by a manipulator, the ultrasound focal point was scanned all over the rubicin combined with FUS treatment (Dox-FUS), Doxorubicin loaded biofilm with 1 mm gap. The spatially- and temporally-averaged acoustic microbubble complex (MB-NP-Dox) only, and MB-NPDox combined power was measured using the radiation force method to be 60 W/ with FUS treatment (MB-NP-Dox-FUS). Animals were treated on a cm2by the measured acoustic power divided the surface area of the weekly basis for three weeks and post-treatment monitoring was transducer. Experiments on a control group were conducted, following followed for five weeks. the same procedure but keeping the ultrasound off. Additional control RESULTS As results, therapeutic effects of MB-NP-Dox-FUS will be was done by repeating the experiments using gentamicin solution alone presented as well as the side effects of each treatment to address without liposomes. All treatments were repeated at least three times. the safety of using MB-NP-Dox. RESULTS Bacterial colony forming units were measured. The four dif- CONCLUSIONS Based on this result, further study will be followed to ferent cases include: antibiotic alone shown in Fig. 1 A; ultrasound improve therapeutic effects of the combined treatment as well as alone US shown in Fig. 1 B; ultrasound plus the antibiotic drug in so- the drug loading efficiency of MB complex. lution shown in Fig. 1 C; and focused ultrasound and the antibiotic encapsulated in liposomes shown in Fig. 1 D. CONCLUSIONS Our experimental results have shown that the fo- P49 cused ultrasound can burst the liposomes and release the drugs. Its killing effects were very close to the case that uses the detergent to Bio-film mitigated by drug loaded liposomes promoted by pulsed lyse the liposomes. The focused ultrasound plus gentamicin-encapsu- ultrasound 1 2 Junru Wu , Faqi Li lated liposomes (D) reduce the number of viable bacteria residing 1 2 University of Vermont, Burlington, Vermont, USA; Chongqing Medical inalginate based biofilm by 72 percent (p<0.001). Acknowledgment: University, Chongqing, China Funded by NASA (Cooperative agreement number NNX13AD40A) & Correspondence: Junru Wu National Natural Science Foundation of China (Grant Nos. 81127901, Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P49 11574039,81201102, 11274404). Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 85 of 122 P51 Multi-stage surface acoustic wave for separation of cancer cells in microfluidic device Kaiyue Wang, Wei Zhou, lili niu, Feiyan Cai, Fei Li, Long Meng Shenzhen Institutes of Advanced Technology, Shenzhen, China Correspondence: Kaiyue Wang Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P51 OBJECTIVES The separating of circulating tumor cells (CTCs) is significant in the early diagnosis, prognostic judgement, and provides an effective evidences for guideness of clinical chemotherapy. In order to detect CTCs effectively, many separation techniques have been developed to date, in- cluding functionalized polymers, pinched flow fractionation, hydro- dynamic filtration, inertial microfluidics, deterministic lateral displacement, fluorescent activated cell sorting. With the advantages of non-contact and high throughput, acoustic manipulation has received increasing attention in the cell separation. To ensure the focused cells at a certain position relative to the microchannel, it is essential to align the piezoceramic trans- ducer (PZT) and microchannel precisely, which affects the stability and re- liability of the device. In this paper, a microfluidic device with two Fig. 1 (abstract P49). A statistical analysis was performed using the acoustic stages has been developed to concentrate and separate the can- pairwise comparison “FTest Two Samples for Variances” method. The cer cells without the need of the precise alignment and sheath flow. number of viable bacteria in the alginatebased biofilms were METHODS The standing surface acoustic wave (SSAW) device including a reduced by 72 percent. No statistical difference (p>0.05) was pair of straight interdigital transducers (IDTs) and a pair of circular IDTs found among A, B, C cases was fabricated on the surface of a 128°<span style="font-size:12px; line- height:19.2px"> </span>Y-rotated, X-propagating LiNbO3 substrate. The straight IDTs were used to generate SSAW for cell concentration and the advantages of circular transducers were exploited to generate TSAW for cell separation. The polydimethylsiloxane (PDMS) microchannel was bonded to the SSAW generator using oxygen plasma treatment. And a sample of the mixture of U87 cancer cells and RBCs was used to demon- P50 strate the manipulations of concentration and separation. The frequencies Preparation and characteristics of targeted phase-shift lipid of continuous signals and pulse signal sent to the straight IDTs and the fo- nanoparticles mediated by tumor homing and penetrating peptide cused IDTs were set to be29.74 MHz and 38.74 MHz, respectively (Fig. 1). Leilei Zhu RESULTS We examined the ability of the device by manipulating micro- Ultrasound Imaging, Chongqing, China spheres with about 2μm diameter. When the stream flowed in SSAWs, the Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P50 microspheres promptly arranged in the center of the microchannel. By modulating the relative phase, △Φ, between two IDTs, the position of the OBJECTIVES To prepare a novel ultrasound contrast agent, microspheres relative to the microchannel could be adjusted arbitrarily. targeted phase-shift lipid nanoparticles mediated by tumor When the concentrated cells passed through TSAWs, the microspheres homing and penetrating peptide tLyP-1, and to evaluate its were accurately separate into two categories under theaction of radiation characteristics. force generated by TSAWs, and thus the microspheres separation could METHODS The nanoparticles were prepared by filming-rehydration be achieved. and acoustic-vibration methods. The morphology, distribution, par- CONCLUSIONS The experimental results reveal that the multi-stage ticle size and zeta potential were detected. After heating and irradiat- acoustic-based approach utilizing microfluidic device a promising tech- ing of low intensity focused ultrasound (LIFU), the phase-shift nique in effectively realizing the concentration and separation of cells. characteristic and the enhancement effect in vitro were observed. This kind of acoustic-based devices can avoid the cellular damage caused The tumor homing and cell-penetrating properties of the nanoparti- by shear force which is generated by water dynamics. In addition, the cles were examined by confocal laser scanning microscopy and flow use of this method can avoid precise alignment of microchannel and cytometry. The cytotoxicity of the nanoparticles was evaluated by IDTs, improving the stability and practicability of the system. CCK8 assay. RESULTS Thesizeand distribution of nanoparticles were uni- formed. The size and zeta potential of nanoparticles were (399.50±29.98) nm and (3.28±1.72) mv, respectively. When the nanoparticles were heated to a temperature of 45 °C or after ir- radiated by LIFU, nanoparticles generated phase-shift and en- hanced ultrasound imaging in vitro(P<0.05). The confocal laser scanning microscopy showed that nanoparticles can targetedly aggregate to cell membrane of MDA-MB-231 and penetrate into the cell, but not to HUVEC. The flow cytometry showed that intracellular fluorescence intensity of MDA-MB-231 was higher than that of HUVEC(P<0.05). The CCK8 assay indicated that dif- ferent concentrations of nanoparticles had no significant effects on cell activity (P>0.05). CONCLUSIONS A novel ultrasound contrast agent, targeted phase-shift lipid nanoparticles mediated by tumor homing pene- trating peptide tLyP-1, was prepared successfully. It can target to MDA-MB-231 cell and penetrate into the cell in vitro,and en- Fig. 1 (abstract P51). Schematic diagram of experiment. 1. The hance ultrasound imaging in vitro after LIFU irradiation, which microspheres injected into the microchannel. 2. The microspheres expected to be a novel tumor targeted ultrasound contrast align in the center of the microchannel by the SSAW 3. The microspheres agent and achieve ultrasound molecular imaging at the level of were driven to the upper wall of the channel by the focused TSAW tumor cell. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 86 of 122 P52 Magnetic nanoliposomes as in situ microbubble bombers for multimodality image-guided cancer theranostics Yang Liu, Fang Yang, Chuxiao Yuan, Mingxi Li, Tuantuan Wang, Ning Gu School of Biological Science and Medical Engineering, State Key Laboratory of Bioelectronics, Jiangsu Key Laboratory for Biomaterials and Devices, Nanjing, JiangSu, China Correspondence: Yang Liu Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P52 Fig. 2 (abstract P52). Fabrication and characterization of AMLs and size distribution of liposomes, ALs, and AMLs, respectively OBJECTIVES The physiological barriers imposed by the abnormal tumor vasculature and the dense collagen matrix have prevented nanocarriers from being delivered to unperfused deep tumor regions, which results in weakening the effectiveness of cancer therapeutics. Herein, we designed a nanoliposome drug delivery system aimed at the improvement of nanocarrier accumulation and distribution in the tumor’s poorly accessible regions for excellent diagnosis and potential therapeutic effects for imaging-monitored accurate tumor ablation. METHODS The anethole dithiolethione (ADT) loaded magnetic nano- liposome (AML) delivery system was prepared by a hydration and membrane-filtering method, which consists of ADT, hydrogen sulfide (H2S) pro-drug, doped in the lipid bilayer, and superparamagnetic nanoparticles (MNPs) encapsulated inside (Fig. 1). The size distribu- tion and morphology of the liposomes were characterized (Fig. 2). The generation of H2S gas in HepG2 cells was investigated by a real- time live cell optical imaging system (Figs. 3, 4). High-resolution MRI (7T) and microbubble-enhanced US imaging was performed pre- and post-injection of AML. The therapeutic efficiency of the AML in the tumor-bearing nude mouse model was also evaluated. RESULTS The presence of MNPs trapped and dispersed in the core of the liposome (about 200 nm) was confirmed. The optical microscopic images of a Hep G2 cell showed when incubated with AMLs, a change in cellular morphology and even cell rupture was observed with the in- crease of incubation time. After 6h of incubation, the cells were dis- rupted and detached from the dish bottom. For in vivo applications, when intravenous injection of AMLs, the targeting of AMLs to tumor was enhanced under exposure to an external magnetic field. Time- dependent in vivo MR and US imaging of tumors in a HepG2-bearing mouse model was significantly enhanced. Moreover, after 7-day follow- up observation, AMLs with magnetic field treatments have indicated extremely significantly higher inhibitions of tumor growth. CONCLUSIONS In summary, AMLs are feasible as both synergistic Fig. 3 (abstract P52). Live cell optical system for observation of cell agents to strengthen tumor ablation efficiency and dual-mode con- morphology change after incubation with liposomes, ALs, and AMLs. trast agents to provide significant contrast enhancement for MR and For each sample, a time gradient was acquired ultrasound imaging. This proposed strategy with both enhanced tumor accumulations of liposomes as well as the property of nano- particles to microbubbles conversion holds great promise for multi- modal image-guided accurate cancer therapy. Fig. 4 (abstract P52). Timedependent in vivo MR, US and NIRfluorescence imaging of tumors after intravenous injection of Fig. 1 (abstract P52). Concepts and schematics of AMLs and their samples in a HepG2bearing mouse model, and histological analysis nano to micro conversion for US/MR dual modal imaging and the of excised tumors and major organs after AMLs with and without spatiotemporal bombed combination tumoraccurate therapy magnetic field (MF) treatment Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 87 of 122 P53 OBJECTIVES To prepare a hematoporphyrin monomethyl ether (HMME)- Incubation of free bubble water and phospholipids to prepare loaded poly (lactic-co-glycolic acid) (PLGA) microcapsules (MBHMME/ shelled nanobubbles ultrasound contrast agents PLGA), which could not only function as efficient contrast agent for ultra- Jilai Tian, Fang Yang, Juan Jin, Ning Gu sound (US)/photoacoustic (PA) imaging, but also as a synergistic agent Biological Science and Medical Engineering, Southeast University, for high intensity focused ultrasound (HIFU) ablation for its sonodynamic Nanjing, Jiangsu, China therapy(SDT). Correspondence: Jilai Tian METHODS MBHMME/PLGA was prepared with the double emulsion Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P53 evaporation method by Sonosensitizer HMME nanoparticles were in- tegrated into PLGA microcapsules. After characterization, the cell-kill- OBJECTIVES Nanobubbles less than 1 μm could make a promising ing and cell proliferation-inhibiting effects of MBHMME/PLGA application in ultrasound molecular imaging and drug delivery. How- microcapsules on ovarian cancer SKOV3 cells were assessed. The ever, the fabrication of stable gas encapsulation nanobubbles is still characteristic fluorescence peak of FCLA free acid (reactive oxygen challenging. Free gas bubbles without any shell materials may have species (ROS)-based chemiluminescence reagent) at around 520 nm lots of applications such as water purification, drag reduction, gener- was detected with the fluorescence spectrophotometer. The US/PA ation of free radicals and so on. Herein, we try to make lipid shelled imaging-enhancing effects and synergistic effects on HIFU were eval- nanobubbles with free gas bubble water. uated both in vitro and in vivo. METHODS Phospholipids dried under vacuum or the lyophilized RESULTS MBHMME/PLGA were highly dispersed with well-defined phospholipids were employed. The characters of size, size distribu- spherical morphology (462 ± 0.52 nm in diameter, PDI = 0.932). En- tion, surface tension, viscosity and their ultrasound imaging in vitro capsulation efficiency and drug loading efficiency were 58.33 ± of the prepared lipids-shelled nanobubbles were investigated. The 0.95% and 4.73 ± 0.15%, respectively. The MBHMME/PLGA remark- mechanism of the incubation was also discussed. ably killed the SKOV3 cells and inhibited the cell proliferation, signifi- RESULTS Results show that this type of GU-Liposome has mean cantly enhanced the US/PA imaging results and greatly enhanced diameter of 194.4± 6.6 nm and zeta potential of -25.2± 1.9 mV with the HIFU ablation effects on ovarian cancer in nude mice by the layer by layer self-assembled lipid structure. The acoustic imaging HMME-mediated sono-dynamic chemistry therapy (SDT). analysis in vitro indicated that ultrasound imaging enhancement CONCLUSIONS MBHMME/PLGA represents a potential multifunc- could be acquired by both perfusion imaging and accumulation im- tional contrast agent for tumor diagnosis and treatment, which aging. The dispersed phospholipid molecular in the prefabricated might provide a novel strategy for thehighly efficient imaging-guided free nanobubbles water was expected to be assembled to form con- non-invasive HIFU synergistic therapy for cancers by SDT in clinic. trollable stable lipid encapsulation gas containing ultrasound-sensi- The mechanism of singlet oxygen could be generated from tive liposome (GU-Liposome). Compared with conventional HMMEand MBHMME/PLGA irradiated with HIFU needed further in- mechanical agitation methods, pre-prepared free gas bubble-based vestigation in the future. nanobubbleshave exhibited the controllable nano-size, lower polydis- persity index. CONCLUSIONS Thus, by incubation of free bubble water and phos- P56 pholipids, stable lipid-shelled nanobubbles could be prepared to Preparation of phase-transition perfluoropentane nanodroplets broaden the biomedical application of nanobubbles in the theranos- modified by folate for ultrasound molecular imaging in ovarian tics in future. cancer Jianxin Liu Chongqing Medical University, Chongqing, China P54 Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P56 Compare ultrasound-mediated cavitation between flowing phase- shift nanodroplets and lipid-shelled microbubbles during focused OBJECTIVES This study aimed to develop a hybrid platform based on ultrasound exposures folate-modified phospholipid-shell and perfluoropentane nanodro- Siyuan Zhang, Tianqi Xu, Zhiwei Cui, Sihao Liu, Dapeng Li, Rui Guo, plets (FA-NDs), which could in vitro and in vivo target ovarian cancer Junjie Wang, Yujin Zong, Mingxi Wan and enhance ultrasound imaging after acoustic droplet vaporization Department of Biomedical Engineering, Xi'an Jiaotong University, Xi'an, (ADV) induced by low-intensity focused ultrasound (LIFU). Shaanxi, China METHODS The nanodroplet was fabricated with HSPC, DSPE-PEG Correspondence: Siyuan Zhang (2000) folate, DPPG, cholesterol and perfluoropentane using lipid film Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P54 hydration method and rotary evaporation method. The nanodroplet stability was evaluated at 4 and 37 respectively. The in vitro targeted ef- This abstract is not included as it has already been published: ficiency were tested with SKOV3 cells and in vitroADV was appraised in Zhang S, Cui Z, Xu T, Liu P, Li D, Shang S, Xu R, Zong Y, Niu G, Wang jellium model with LIFU. The in vivo targeted efficiency and acoustic S, He X, Wan M. Inverse effects of flowing phase-shift nanodroplets droplet vaporization were evaluated with SKOV3 tumor-bearing nude and lipid-shelled microbubbles on subsequent cavitation during fo- mice. cused ultrasound exposures. Ultrason Sonochem. 2017; 34: 400-409. RESULTS The nanodroplets were successfully prepared with good size Available from: http://www.sciencedirect.com/science/article/pii/ uniformity (particle size 321±67 nm). The nanoparticles remained stable S1350417716302139. for 48 h at 4 and 1 h at37. In vitro targeted experiments exhibited a perfect binding efficiency of FA-NDs to SKOV3 cells. In vitro ADV pro- files displayed obvious ultrasound enhancement in both B-mode and P55 CEUS-mode when LIFU power was elevated to 5 W. In vivo and ex vivo Sonodynamic Therapy (SDT) polymer contrast agent for fluorescence imaging displayed that FA-NDs possessed outstanding ultrasound/photoacoustic dual-modality imaging-guided specificity to targeted solid tumor. Both the qualitative and quantitative synergistic High Intensity Focused Ultrasound (HIFU) therapy results of in vivo ADV in mice tumor nodule manifested that FA-NDs Lan Hao underwent phase transition upon the LIFU exposure. Chongqing Key Laboratory of Ultrasound Molecular Imaging, the CONCLUSIONS The results demonstrated that the FA-NDs system is a Second Affiliated Hospital University, Chongqing, China potential platform for targeted conjunction and enhancing ultra- Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P55 sound imaging via ADV using LIFU. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 88 of 122 P57 ited good photostability. Upon near-infrared laser irradiation, the PLGA- Targeted Pegylated PLGA coated prussian blue nanocomposite for PB-PTX-PEG-FA nanocomposite showed an enhanced synergistic photo- dual-modality PA/MR imaging and synergistic chemo-thermal thermal and chemical therapeutic efficacy forbreast cancer than solo tumour therapy photothermal therapy or chemotherapy. Tingting Shang CONCLUSIONS In conclusion, we prepared highly dispersed core- Institute of Ultrasound Imaging Department of Ultrasound, The Second shell structure PLGA-PB-PTX-PEG-FA nanocomposite by coating PB Affiliated Hospital of Chongqing Medical University; Chongqing Key NPs with PLGA-FA shell and then modifying with PEG. The prepared Laboratory of Ultrasound Molecular Imaging, Chongqing, China. PLGA-PB-PTX-PEG-FA nanocomposite has good biocompatibility, ex- Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P57 cellent photo-thermal transformation capacity, and can enhance both PA imaging and MR imaging in vitro and in vivo. Furthermore, PLGA- OBJECTIVES To prepare a PEGylated poly (lactic-co-glycolic acid) PB-PTX-PEG-FA nanocomposite acts as a multifunctional drug deliv- (PLGA) targeting with folic acid (FA) coated Prussian blue nanoparti- ery system with higher drug loading capacity and excellent drug cles (PB NPs) and paclitaxel(PTX), to construct multifunctional PLGA- controlled release-system. Finally, our result confirmed that PLGA-PB- PB-PTX-PEG-FA nanocomposite for both photoacoustic (PA) imaging, PTX-PEG-FA nanocomposite has the ability of the synergistic thera- magnetic resonance imaging (MRI) and synergistic chemo-thermal peutic efficacy combined photothermal and chemical therapy effect, tumor therapy. which further enhances the therapeutic efficacy to breast cancer. METHODS Paclitaxel (PTX)-loaded Folic acid (FA) targeted PEGylated PLGA nanoparticles encapsulating Prussian Blue (PB) (PLGA-PB-PTX-PEG- FA) nanocomposite was fabricated by a modified double emulsion (water/oil/water) evaporation process. The morphology, size, UV–vis–NIR P58 absorbance spectra, Fouriertransfer infrared (FTIR) spectrum were tested Low intensity focused ultrasound regulates drug release from to evaluate the structural characterization of PLGA-PB-PTX-PEG-FA. Drug porphyrin-phospholipid liposomes and facilitates multi-functional Loading and releasing of PLGA-PB-PTXPEG-FA nanocomposite were theranostics 1,2 1, 2 1 1 assessed by high performance liquid chromatography (HPLC). Addition- Xiaobing Wang , Xiufang Liu , Fei Yan , Hairong Zheng ally, in vitro cell targeting and in vivo tumor targeting were verified by Shenzhen Institutes of Advanced Technology, Chinese academy of confocal laser scanning microscopy (CLSM) and vital fluorescence im- Sciences, Shenzhen, China; Shaanxi Normal University, Xi'an, China aging (VFI) to assess the targeting effect of PLGA-PB-PTX-PEG-FA nano- Correspondence: Xiaobing Wang composite. The efficacy of the contrast agent for PA imaging and MRI Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P58 was evaluated by in vitro and in vivo imaging of subcutaneous MDA-MB- 231 tumor-bearing mice following tail intravenous injection of the con- OBJECTIVES Nanoscale drug delivery systems (DDS) facilitate multifunc- trast agent. Finally, to visualize the photothermal cytotoxicity of PLGA- tional theranostics. Externally controlled drug release from DDS provides PB-PTX-PEG-FA nanocomposite, MDA-MB-231 cells were incubated with selective targeting. Here, we introduce an ultrasound-activatable porphy- PLGA-PB-PTX-PEG-FA nanocomposite in 96 well-plate for 2 h and were rin-phospholipid liposome (PPL) to show its bimodal imaging, robust car- irradiated with or without NIR laser. CCK8 method was used to measure rier, drug release behavior and efficient therapeutic actions. the viability of the cells. To study the photothermal effect and synergistic METHODS Dynamic light scattering and transmission electron micros- therapeutic efficacy, PLGA-PB-PTX-PEG-FA nanocomposite suspensions copy were utilized to characterize the features of PPL and PPL loaded of various concentrations nanoparticle dispersions were exposed to a with doxorubicin (PPLDOX) (Fig. 1). The ultrasound controllable drug re- laser with the wavelength of 808 nm and the output power of 0.647W lease was measured at distinct time points with or without different for 10min in vitro or in vivo after vein intravenous injection. free radical inhibitors (Fig. 2). In vitro cellular uptake and cytotoxicity RESULTS The highly dispersed PLGA-PB-PTX-PEG-FA nanocomposite were examined in U87 glioma cells by confocal microscopy and micro- with spherical morphology with smooth surface was directly observed by plate reader. In vivo, photoacoustic and NIR fluorescent imaging were TEM and SEM and has good uniformity with an average hydrodynamic applied to indicate the distribution of PPL following intravenous injec- diameter of 236.6±55.04nm. That PB NPs being encapsulated by PLGA- tion, along with the signals change after ultrasonic irradiation (Fig. 3). FA-PEG was confirmed by TEM, UV–vis–NIRabsorbance spectra and FTIR The tumor inhibition and overall survival timewere evaluated by PPL- spectrum of the PLGA-PB-PTX-PEG-FA and PB NPs. Both PB and PLGA- DOX endowed with the optimal porphyrin-phospholipid ratio, max- PB-PTX-PEG-FA nanocomposite displayed a broad absorption band from imum DOX loading and favorable ultrasound-responsible ability (Fig. 4). 500 nm to 900 nm with a strong absorption peak at ~702 nm. While the RESULTS The obtained PPL, PPL-DOX were stable at 4°C for at least two IR spectrum of the PLGA-PB-PTX-PEG-FA nanocomposite showed both months with appropriate 100 nm diameter. The ultrasound induced drug the corresponded peaks of PLGA-FA-PEG and PB NPs. The results of drug release from PPLDOX increased with higher molar phorphyrin-phospho- loading efficiency and drug loading capacity tested by HPLC were lipid, while the drug loading efficient declined as phophyrin-phospholipid 77.82% and 7.22%. The fast drug release from the nanocomposite in vitro ratio increased. 2 molar % porphyrin-phospholipid in PPL-DOX was com- with laser irradiation indicated the great potential as a controlled-release patible with good ultrasound-responsible and drug-loading capacities. system for the anticancer drug. In vitro cell targeting and in vivo tumor Under this condition, DOX release from PPL-DOX could beregulated by targeting of PLGA-PB-PTX-PEG-FA nanocomposite, targeted group had ultrasound intensity and abolished with different free radical inhibitors. Fol- obvious difference from non-targeted group and blank control group, lowing intravenous administration, the PPL demonstrates high sensitive which showed that good targeting effect to tumor cells or tumor in vivo photoacoustic and NIR fluorescent imaging and effective tumor site drug of PLGA-PB-PTX-PEG-FA nanocomposite. After the tail intravenous injec- delivery in xenograft U87-bearing mice. Ultrasound exposure triggers sono- tion of thePLGA-PB-PTX-PEG-FA nanocomposite, the PA images and MR dynamic damage of tumors cells and simultaneously initiates local drug re- images of tumor of nude mice were significantly enhanced, while there lease to exert chemotherapy. Exposure to focused ultrasound with PPL were almost no distinct enhancement in non-targeted group and blank suppresses tumor growth several times more than without exposure to control group. The result of the photothermal cytotoxicity of PLGA-PB- ultrasound. What’s more, PPL-DOX displays little damage on normal cells PTX-PEG-FA nanocomposite showed that the viability of the cells of in vitro and even on normal tissues in vivo. PLGA-PB-PTX-PEG-FA nanocomposite+Laser group was lowest. The re- CONCLUSIONS The developed ultrasound-activable paradigm sults of photothermal conversion property of PLGA-PB-PTX-PEG-FA sug- achieves simultaneous photoacoustic/fluorescence imaging and gested the character of photothermal effect was positively correlated spatiotemporally regulates nano-drug release and initiates sono- with the heating power and the nanoparticles concentration and exhib- chemotherapeutic effects. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 89 of 122 Fig. 1 (abstract P58). PPL-DOX characteristics. (a) Transmission electron microscopy of PPL-DOX liposome morphology under Fig. 4 (abstract P58). In vivo NIR flourescent imaging. (a) The different magnifications with 2% Pyro-lipid (Molar ratio). distribution of drug in tumor after I.V. injection. (b) Ultrasound (b) Particel size analysis of PPL-DOX with different Mol % comnbined with microbubbles enhanced the fluorescence of porphyrin-phospholipid Pyro-liposome, facilitating NIR and photoacoustic imaging and sonochemical therapeutics. A dual tumor model was used, with a tumor on the right flank was irradiated and the other was non-irradiated P59 Protective effects of static magnetic field on the sonoporation-induced cellular ionic imbalance Yaxin Hu, Mei Yang, Yancheng Wang, Siping Chen Biomedical Engineering, Shenzhen University, Shenzhen, Guangdong, China Correspondence: Yaxin Hu Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P59 OBJECTIVES During the past two decades, great efforts have been made to investigate the perforation and recovery dynamics of sono- poration-created pores on the plasma membrane (PM). Although both Fig. 2 (abstract P58). Ultrasound triggered drug release from acoustic and non-acoustic parameters have been extensively optimized PPL-DOX with (a) or without (b) free radicals scavengers to facilitate the pore-resealing process, sonoporation mediated intracel- lular delivery is still challenged by the fact that a significant proportion of the sonoporated cells would undergo apoptosis. It has been hypoth- esized in previous investigations that excessive ion exchanges across the PM pores could result in a cellular ionic imbalance, which might consequently activate the apoptosis signaling pathway. To provide dir- ect evidence for this hypothesis, this work will study the ion exchange dynamics in sonoporation and its potential relevance to apoptosis in- duction. Particularly, our work will also examine the question of whether the ion exchange dynamics in sonoporation could be modu- lated by a magnetic field. METHODS To investigate the ion exchange dynamics across the PM pores, a real-time sonoporation experiment platform was established. More specifically, an adherent microbubble (Targestar® P) was firstly intro- duced to the apical PM of a single MRC-5 cell. Then, inertial cavitation of the microbubble was triggered by a single ultrasound pulse (center fre- quency: 1 MHz; peak negative pressure: 0.85 MPa; pulse duration: 10 μs) and monitored by a high-speed camera (imaging rate of 680,000fps). In this way, the ion exchange dynamics in sonoporation could be imaged by a high-resolution confocal microscope (LSM 710). To record the cal- cium influx insonoporation, a green-fluorescent dye (Fluo 4, AM) was in- troduced to the cytoplasm. To evaluate the mitochondria damage in sonoporation, a red-fluorescent dye(TMRE) was used to indicate the mito- chondria membrane potential (MMP). Nucleic acid stains of Sytox Blue Fig. 3 (abstract P58). The characterization of PPL flourescene. and PI were added to the sonoporated cells at different timepoints (0 s (a) Flourescene emission of intact quenched PPL (in PBS) versus and 10 min post-sonoporation, respectively) to examine the efficiency of disrupted unquenched PPL. (b) Flourescene intensity of PPL upon PM resealing with and without a static magnetic field (680 mT). Qualita- ultrasound irradiation range from 0.25 Mpa to 0.35 Mpa. (c) Flourescene tive andquantitative analyses of fluorescent images were carried out photographs of PPL post ultrasound irradiation was recorded using the ImageJ software. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 90 of 122 RESULTS As illustrated by the green-fluorescent images in Fig. 1, cal- biofilms. This part of the study was conducted according to the method cium influxes (indicated by red arrows, top row) into the sonopo- described by Singh et al. rated cell (cell #2) could be found at the microbubble-adherent sites RESULTS Bacterial morphology. In group C, bacteria within biofilms (2 s post-sonoporation). The intracellular calcium concentration of were closely linked and wrapped by abundant fiber-like materials, caus- the sonoporated cells increased to the peak value at 4 s post sono- ing the formation ofhighly organized multicellular population structure poration, and the mitochondria membrane potential greatly changes (Fig. 1A). In group U, however, less fiber-like material was observed in at this time point (red fluorescence, bottom row). In the presence of biofilms structure. Some bacterial debris, however, was visible and the static magnetic field, both the ion influx speed and the MMP change bacterial surface exhibited deformation (Fig. 1B and C). Bacterial viability. amplitude were reduced. With respect to the PM resealing, the mag- Bacterial plate counting was performed to evaluate bacterial viability. netic field increased the percentage of cells with successful PM re- Compared with group C, bacterial viability in groups U (P=0.002), covery by 17.3 % (n=38). G(P=0.003), and T (P=0.001) were significantly decreased. Moreover, bac- CONCLUSIONS Compared with the cellular organelles that are an- terial viability was even more significantly decreased in groups U+G chored to the cytoskeleton, the freely diffusible ions of the cytoplasm (P=0.001) and U+T(P=0.001) (Fig. 2). According to Formulation 1, groups are more vulnerable to the sonoporation-induced disturbance. This U+G and U+T displayed enhanced inhibitory ratios on ESBLs E. coli bio- study directly demonstrated that the calcium influxes across the PM films of 3.7% and 2.4%, respectively. Assessment of ESBLs E. coli biofilms pores could result in cellular ionic imbalance and mitochondria mem- by CLSM. Results were analyzed using the Image J software. In all brane potential changes. This study contributes to show that static groups, the amount of living bacteria was the most in the intermediate magnetic field could protect the sonoporated cells in the scenario of layer, relatively less in the inner layer and least in the outer layer of bio- intracellular delivery of neutrally-charged drugs by reducing the films. In addition, the percentage of dead bacteria in each layer of bio- speed of ion exchanges and promoting PM resealing. films in groupsU, U+G, and U+T significantly increased when compared with the same layer of group C. This was especially true in group U+T (Fig. 3). In order to further observe the differences of biofilms morph- ology and surface viability among these groups, a 3-D surface shape of biofilms was reconstructed with pictures from CLSM. (Fig. 4) Compared with group C, thickness of biofilms in groups U, U+G, and U+T were sig- nificantly decreased (P<0.05), while group AP was significantly increased (P<0.05), indicating increased channel gaps in groups U, U+G, and U+T. Additionally, ADD was significantly decreased in groups U+G and U+T (P<0.05), suggesting the blocked nutrient supply of biofilms in these groups. With the exception of group T, TE decreased significantly in other treatment groups when compared with group C (P<0.05), indicat- Fig. 1 (abstract P59). Sonoporationinduced calcium influxes and ing the decrease of uniformity of E. coli biofilms in groups U+G, U+T, mitochondria membrane potential changes and G. Collectively, these results indicate that biofilms morphology and viability suffer damage after ultrasound treatment, either with or without antibiotics. Antibiotic penetration. To clarify the effects of ultrasound and P60 antibiotics on the permeability of ESBLs E. coli biofilms, a ciprofloxacin Effects of low-intensity and low-frequency ultrasound combined penetration test was conducted using the penetration model. This data with antibiotics on biofilms of Extended-Spectrum suggests that LILFU can increase the permeability of ESBLs E. coli Beta-Lactamases (ESBLs) producing Escherichia coli biofilms to antibiotics. Diameter of inhibition Zone (mm):C:14.0±0.3;G:21.5 1,2 Hexun Jiang ±0.4*;T:23.6±0.4*;U:26.1±0.2*;U+G:31.4±0.1*;U+T:33.6±0.3*(*Compared with Shanghai First Maternity and Infant Hospital, Tongji University School of group C, P<0.05. n=6.) Medicine, Shanghai, China; State Key Laboratory of Ultrasound CONCLUSIONS In conclusion, LILFU enhances the penetrating cap- Engineering in MedicineCo-Founded by Chongqing and the Ministry of ability of antibiotics into ESBLs E. coli biofilms, causing injuries of Science and Technology, State Key Laboratory of Ultrasound bacteria within biofilms. Ultrasound, combined with antibiotics, en- Engineering in Medicine Co-Founded by Chongqing andthe Ministry of hances bactericidal effects against biofilms. Our findings may provide Science and Technology, Chongqing, Chongqing, China a potential therapeutic method for killing bacteria insidebacterial bio- Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P60 films generated by ESBLs E. coli in vitro. OBJECTIVES The purpose of this study is to investigate the effects of low-intensity and low-frequency ultrasound (LILFU), when combined with antibiotics, on bacterial viability and the morphology of anti- biotic penetration associated with biofilms formed by extended- spectrum beta-lactamases (ESBLs) producing Escherichia coli (E. coli), a multidrug resistant bacterium strain. METHODS Experimental groups and ultrasound irradiation. The bio- films were divided into 6 groups: the control (C); gentamicin (G) with a working concentrationof 200 μg/ml; tobramycin (T) with a working con- centration of 200 μg/ml; ultrasound (U); ultrasound combined with gen- tamicin (U+G) with a working concentration of gentamicin of 200 μg/ ml; ultrasound combined with tobramycin (U+T) with a working con- Fig. 1 (abstract P60). See text for description centration of tobramycin of 200 μg/ml.The ultrasound transducer was smeared with ultrasonic coupling gel and attached to the bottom of the 6-well plate. Biofilms of groups U, U+G, and U+T were irradiatedat 42 kHz and 0.66 W/cm2 for 0.5-h, with continuous wave ultrasound. Bacterial plate count. The synergistic effect (SE) was calculated accord- ing to Formulation 1 as follow: Synergy Effect (SE) = the inhibitory ratio in the ultrasound combined with antibiotics group- the inhibitory ratio in the antibiotics group. CLSM analysis. Unattached dye was removed from the biofilms by washing. Signals were recorded using green (exci- tation wavelength at 488 nm and emissionwavelength at 515/30 nm) and red (excitation wavelength at 568 nm and emission wavelength at Fig. 2 (abstract P60). See text for description 600/50 nm) channels. Penetration of antibiotics through ESBLs E. coli Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 91 of 122 Fig. 3 (abstract P60). See text for description Fig. 1 (abstract P61). (A) Upper: The 3D cell morphology P61 reconstruction before and after ultrasound sonication. When mechanisms of intracellular gene delivery via ultrasound and DNAloaded MBs were destructed into fragments, a pore onto cell DNA-loaded microbubbles membrane was generated and the fragments were spread around 1 2 1 Ching-Hsiang Fan , Yu-Chun Lin , Chih-Kuang Yeh the site of pore; lower: another case, the debris was colocalized with Department of Biomedical Engineering and Environmental Sciences, the cell membrane after ultrasound sonication. (blue: cyan National Tsing Hua University, Hsinchu, Taiwan; Institute of Molecular fluorescence protein labeled cell membrane; red: Rhodaminelabeled Medicine, National TsingHua University, Hsinchu, Taiwan DNAloaded MBs) (B) Timelapse images confirmed the gene was Correspondence: Ching-Hsiang Fan transported to cell nuclei from DNAloaded MBs after ultrasound Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P61 sonicaton. (red: Rhodaminelabeled DNAloaded MBs; green: YFPlabeled cell membrane; blue: DAPIdye labeled cell nuclei) OBJECTIVES In the past few years, several studies on ultrasound mediated gene delivery with DNA-loaded microbubbles (MBs) is vastly expanding since it’s noninvasive and non-viral vector properties. Never- P62 theless, little is known on the machinimas of enhanced cellular gene de- Acoustic vaporisation of encapsulated droplets livery following ultrasound with DNA-loaded MBs interaction. In this 1,2 4 3 A. Podkovskiy , C. Olivier4, J.L. Thomas , M. Guedra3, F. Coulouvrat , study, we tried to image the MBs and cells during ultrasound sonication 3 1, 2 2 T. Lacour , Wladimir URBACH , N. Taulier for revealing the short-term and long-term mechanisms of intracellular 1 2 ENS Paris, Lab. Phys. Stat., Paris, France; LIB, Sorbonne Universités gene delivery, including sonoporation, endocytosis, and lipid fusion. UPMC and CNRSLIB, 75005-Paris, France; Sorbonne Universites, UPMC METHODS DNA-loaded MBs were synthesized to have cationic Univ Paris 06and CNRS UMR 7190F-75005 Paris, Institut Jean Le Rond phospholipid shell with C3F8 gas core, with DNA (green fluorescence d’Alembert, Paris, France; INSP, Sorbonne Universités UPMC 75005 Paris, protein plasmid) loaded via electrostatic interactions. The lipid shell of France DNA-loaded MBs was modified with folate acid for binding with the fol- Correspondence: A. Podkovskiy ate acid receptor of C6 glioma cell. The measured mean concentration Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P62 and size of DNA-loaded MBs were (2.1 ± 0.1) × 1010 MB/mL and 1.0 ± 0.2 μm, respectively, measured by multisizer. The DNA loading effi- OBJECTIVES The use of encapsulated droplets is currently largely in- ciency was 27.1 ± 3.1 %. Ultrasound exposure (frequency = 1-MHz, vestigated for medical applications, mainly because their reduced acoustic pressure = 0.5 MPa, cycle number = 500, pulse repetition fre- size allows them to enter targeted areas that cannot be reached by quency = 10 Hz, sonication time = 10s) was performed with presence large microbubbles (contrast agents). Perfluorocarbon droplets can of DNA-loaded MBs to achieve gene delivery with C6 glioma cells. To be vaporized under the action of an ultrasonic field, in order to turn monitor the DNA trafficking from MBs into cell nuclei, the DNA, cell them into echogeneous eventually cavitating microbubbles. The membrane and cell nuclei were fluorescently labelled with Rhodamine, optimization of acoustic droplet vaporization (ADV) will be enhanced yellow fluorescence protein (YFP) and DAPI dye, individually, and im- by understanding the physical mechanisms underlying ADV, which aged by a live-cellmulti-color fluorescent microscope before and after are currently not totally elucidated. ultrasound sonication. During the imaging, the cells were kept in a hu- METHODS A recent model (JASA p3656–3667 (2015) doi:10.1121/ midified atmosphere with 5% CO2 at 37 °C. 1.4937747) describes this phenomenon paying particular attention to RESULTS The 3-D cell morphology reconstruction data show that a pore the finite size of the droplet and its encapsulation by a thin viscoelas- (Fig. 1A, white arrows) on the cell membrane was immediately formed tic layer. Numerical simulations, done for droplets of micrometric when the DNA-loadedMBs (spherical red structure) are disrupted into radii and for frequencies of 1–5 MHz, reveal that droplet surface ten- fragments (red spot) by ultrasound sonication. We also observed that sion and shell rigidity are responsible for an increase of the acoustic the debris of DNA-loaded MBs spread around the site of pore or co-lo- droplet vaporization threshold. This threshold does not vary mono- calized with the cell membrane (white arrows), indicating that the DNA- tonically with frequency and thus an optimal frequency can be found loaded MBs vesicles might through sonoporation or lipid fusion path- to minimize it. To check the above theoretical results, we investi- ways to delivery gene into the cells. The time-lapse images (Fig. 1B) also gated in vitro the relationship between ADV and Inertial Cavitation. confirm that the gene indeed was transported and internalized into cel- RESULTS The threshold pressures required to induce ADV and IC lular nuclei from DNA-loaded MBs following ultrasound sonication. were simultaneously determined for micron-sized PFC droplets as a CONCLUSIONS This study demonstrated the intracellular DNA traf- function of droplet size and US parameters. Experimental conditions ficking from DNA-loaded MBs into cellular nuclei might rely on sono- that yield ADV without IC and ADV with IC, that could enhance drug proation, endocytosis, and lipid fusion. Future works include optimize delivery via sonoporation, are determined by investigating parame- the parameters of ultrasound and DNA-loaded MBs for improving ters that could influence both thresholds; bulk fluid properties such the efficiency of gene delivery. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 92 of 122 as gas saturation, temperature, viscosity, and surface tension; droplet P64 parameters such assurfactant type; and acoustic properties such as DOX-loaded superhydrophobic mesoporous silica nanoparticles pulse repetition frequency and pulse width. with ultrasound for dual-treatment in tumour CONCLUSIONS In conclusion, our experiments allowed us to test a QiaoFeng Jin, Cheng-Han Wu, Chih-Kuang Yeh new model of droplets vaporization using ultrasound in an infinite Department of Biomedical Engineering and Environmental Sciences, medium. The study of droplets vaporization, located in confined National Tsing Hua University, Hsinchu, Taiwan medium more resembling the situation in vivo, will be the next step. Correspondence: QiaoFeng Jin Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P64 OBJECTIVES Mesoporous silica nanoparticles (MSNs) have great P63 potential for biomedical applications. Surface air bubbles Ultrasound theranostics for tumor adsorbed on the fluorine-modified superhydrophobic MSNs could Kazuo Maruyama be served as cavitation nuclei and prevent the leakage of chemo- Faculty of Pharma-Sciences, Teikyo University, Tokyo, Japan therapy drugs encapsulated into MSNs during circulation. Com- Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P63 bination of superhydrophobic surfaces with ultrasound can cause cavitation to induce cell damage and facilitate drugs release. In OBJECTIVES “Theranostics” is a treatment strategy that combines this study, we propose a dual-treatment strategy by combining therapeutics with diagnostics. The combination of bubble formula- active cavitation and chemotherapy. tion and ultrasound (US) is a good tool for “theranostics” due to have METHODS Synthesis of superhydrophobic F48-ext and Dox loaded multi-potency both of diagnostics with enhanced echo signal from F48-extThe parent MCM-48 type MSNs were synthesized with sol- bubble and therapeutics with cavitation of bubble. To develop a gel method using BCDAC and C16E2 as soft templates. To fabri- novel bubble formulation for US imaging and therapy with small par- cate fluorinated MCM-48 (names as F48-ext), the MCM-48 were ticle size and a good stability and test the formulation as US imaging dispersed in 10 mL toluene containing 0.2 ml of perfluorodecyl- contrast agent and for gene delivery in vitro and in vivo. Interleukin12 triethoxysilane and stirred at 100° for 48 hours, and the resulted (IL-12) exhibits immunomodulatory antitumor effects and is consid- solid was collected and dried at 60° for 12 hours. Finally, F48-ext ered an effective antitumor agent, but its short half-life and systemic NPs were obtained after removing the surfactants by repeated toxicity following intravenous injection are major obstacles to its ion exchange in a dilute HCl-ethanol solution at 60°. The DOX therapeutic use. Therefore, we transfected pDNA encoding the IL-12 was loaded into F48-ext NPs in ethanol which contains 2% (v/v) gene (pCMV-IL-12) into tumor tissue using bubbles and US with the concentrated hydrochloric acid, then the ethanol were vaporized aim of achieving high local expression of IL-12. and replenished with 1ml ethanol contain 2% hydrochloric for METHODS Lipid-stabilized bubbles were prepared by homogenization thrice. Inertial cavitation dose measurement using a passive cavi- of a lipid dispersion in the presence of perfluoropropane gas. Different tation detectorA 15-MHz focused transducer was used as a pas- phospholipid compositions were tested and evaluated. After bubble sive cavitation detector to receive inertial cavitation (IC) signals, formation the bubbles were freeze-dried so that a dry sample contain- while a 2-MHz HIFU transducer was used to excite theNPs. The ing bubbles was formed. After re-constitution of the samples they were integrated value of the amplitude of frequency spectrum from analyzed for size, gas content and US signal intensity. The ultrasound 10-20 MHz was termed inertial cavitation dose (ICD) and used to theranostics capabilities of bubbles for the solid tumor were studied in assess the IC intensity. In vitro cell cytotoxicity with alamar Blue® Colon26 tumor bearing mice. Bubbles was injected to mice via tail vein with and without US exposureTRAMP cells were incubated with and 9 MHzlinear ultrasound was exposed to solid tumor site transder- Dox-loaded F48-ext NPs, free F48-ext NPs and free Dox for 2 mally. Following the recognition of neovasculature in tumor tissue, 1 hours in 24 wells plate, respectively and then all samples were MHz therapeutic ultrasound was exposed transdermally over the site of sonicated at 5 MPa pressure with a PRF of 100 for 5 min by the solid tumor tissue. 2-MHz transducer. The cells without ultrasound sonication were RESULTS Bubbles was injected to mice via tail vein and 9 MHz linear set as control group. After 3 hours, the samples were washed ultrasound was exposed to solid tumor site transdermally. The flow and further incubated for 24 hours, and their cytotoxicity were of bubbles in blood was observed and neovasculature of tumor tis- measured by using alamar Blue® kit. sue was imaged clearly. Following the recognition of neovasculature RESULTS MCM-48 and F48-ext NPs show average sizes of 200-300 in tumor tissue, 1 MHz therapeutic ultrasound was exposed transder- nm, and show the zeta potentials of -30 mV, which contributed to mally over the site of solid tumor tissue. This process induced cavita- the stability of such NPs. The F48-ext NPs had a contact angle of tion of bubbles in the tumor tissue, resulted in rising the about 150°, and after loading Dox their contact angle decreased to temperature of tumor tissue to 45-55C, and also significant reduction 130° due to the hydrophilic Dox. The ICD of F48-ext NPs have signifi- of tumor growth. Cavitation leads to localized heating and cloud be cantly augmented with respect to that of MCM-48 NPs (Fig. 1). The use for ablative cancer therapy. Transfection of pCMV-IL-12 with LBs ICDs of F48-ext and Dox-loaded F48-ext NPs both increased with and US suppressed tumor growth significantly. To investigate the acoustic pressures. Due to the superhydrophobic modification, the mechanism behind the anti-tumor effects of pCMV-IL-12 transfected Dox leakage of F48-ext NPs was suppressed. We found that Dox re- using bubbles and US, we assessed the involvement of CD4+ and leasing of F48-ext NPs was not increased with ICD increasing. For the CD8+ T cells and NK cells. The depletion of CD8+ T cells effectively short term, the cells damage caused by inertial cavitation of F48-ext blocked the antitumor effect of pCMV-IL-12 transfected using bub- NPs was observed and dominated the cell cytotoxicity. The Dox re- bles and US. These results suggest that the combination of bubbles leasing from the F48-extNPs performed long term treatment mode. and US can effectively induce sufficient IL-12 expression to cause CONCLUSIONS In summary, the degree of inertial cavitation of MSNs anti-tumor immune responses. increased after fluorine-modification (F48-ext), and the Dox-loaded CONCLUSIONS The combination of bubbles and US could be effica- into F48-ext NPs shows better stability than MCM-48 NPs. In addition, cious for neovasculature image and cancer therapy. We believe this DOX loading did not inhibit the IC activity of F48-ext NPs. Combin- new formulation shows great promise for both diagnostic and thera- ation of mechanical effect and chemotherapy have shown the tumor peutic applications thanks to its good stability, relatively small bub- dual-treatment mode. We demonstrated that F48-ext NPs were sensi- ble size and the simplicity of handling. tive to ultrasound and with a stable drug-loading capacity. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 93 of 122 OBJECTIVES To give advice about how to select suitable patients in MR-guided focused ultrasound surgery (MRgFUS) of uterine fibroid by screening pelvic MRI. METHODS Retrospective analysis the MRI image and clinical data of 30 fibroid patients who successfully complete the MRgFUS treatment in Huashan hospital and also review the literature. RESULTS The safe and effective use of MRgFUS is affected by fibroid type and location, position relative to adjacent anatomical structures and the presence of coexistent pelvic disease. CONCLUSIONS Screening pelvic MRI for selection of patients in whom sufficient fibroid volumes can be treated safely using the MRgFUS system is critical forsuccessful outcomes. Fig. 1 (abstract P64). (A) The diagram of controlled drug release P67 from Dox laoded superhydrophobic F48-ext with air layer; (B) the Effect of fiducial markers and brachytherapy seeds on the delivery experimental setup of the ICD measurement; (C) the contact angle of HIFU salvage therapy in the prostate 1 2, 1 3, 1 3 of the MCM-48, F48-ext with and without Dox; (D) the ICD of the Panayiotis S. Georgiou , Jiri Jaros , Heather Payne , Clare Allen , 4 4 1 1 MCM-48, F48-ext and F48-ext Dox; (E) the compare of cell cytotoxicity Taimur T. Shah , Hashim Ahmed , Eli Gibson , Dean Barratt , of F48-ext dox with and without ultrasound exposure Bradley Treeby Medical Physics and Biomedical Engineering, University College London, London, UK; Computer Systems, Brno University of Technology, Brno, CzechRepublic; Oncology, University College London Hospitals, London, UK; Surgery and Interventional Science, University College London, London, UK P65 Correspondence: Panayiotis S. Georgiou The influence of exercise on uterine fibroid and adenomyosis after Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P67 high-intensity focused ultrasound Huang Xueyan OBJECTIVES Prostate cancer is the most common cancer and the sec- Biomedical Engineering, Chongqing, China ond leading cause of cancer-related death in men in Europe and North Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P65 America. For patients with early-stage localized disease, the cancer is often treated using external beam radiation therapy (EBRT) or brachyther- OBJECTIVES To evaluate exercise can improve clinical symptoms of apy. In EBRT the radiation treatment is delivered from a source external uterine fibroid and relief dysmenorrhea of adenomyosis after high-in- to the patient’s body and the procedure usually involves implanting a tensity focused ultrasound (HIFU) treatment. small number of gold fiducial markers into the prostate to verify the pos- METHODS From January 2011 to August 2015. 83 patients ition of the prostate gland between treatments. In brachytherapy the ra- suffered from uterine fibroid and 102 patients suffered from diation dose required for the treatment is delivered locally by implanting adenomyosis. All patients were treated withHIFU.Among the 185 a large number of radioactive seeds (approximately 100). In low dose rate patients, 83 of them completed one-year follow-up,102 patients (LDR) brachytherapy these seeds remain permanently in the patient’s of them completed half a year follow up. All the symptom im- body. For some of these patients, treated either by EBRT or LDR brachy- provement the uterine fibroid volume and the standards of Visual therapy, their cancer will recur. In such cases, further treatment using an Analogue Scale(VAS) were recorded. alternative (salvage) therapy can be considered. High intensity focused RESULTS In this study, 83 patients had uterine fibroid,51 of 83 ultrasound (HIFU) is currently offered in hospitals as a minimally invasive patients were no exercise intervention group, the rest of 83 pa- salvage therapy for treating prostate cancer inpatients whose cancer has tients were exercise intervention group,Comparing with this two recurred. However, clinicians observe mixed results with salvage-HIFU for groups’ uterine fibroid change,P=0.003,considered as statistically failed brachytherapy, which may be partly linked to inadequate heating significant.That meaning, exercise could help improve the absorp- caused by the implanted seeds. To date, the impact of implanted EBRT tion of uterine fibroid.102 patients had adenomyosis,45 of 102 markers and brachytherapy seeds on HIFU treatment has not been thor- patients were no exercise intervention group,57 of 102 patients oughly studied. The objective of this work was to investigate the effect of were exercise intervention group.From the result found that exer- a single EBRT marker on the efficacy of HIFU treatment delivery and ex- cise may relief the score of dysmenorrhea,P=0.002,considered as tend these results to a large number of brachytherapy seeds. statistically significant. METHODS Using the k-Wave acoustics toolbox, we have performed CONCLUSIONS Based on the observations from uterine fibroid coupled acoustic-thermal modelling of HIFU treatments by first model- and adenomyosis, exercise have positive reaction on improving ling the prostate with the presence of a single EBRT gold marker at dif- symptoms of uterine fibroid andadenomyosis. Hence, this clinical ferent positions and then, with the presence of multiple brachytherapy effective and response is a further treatment for woman who seeds. The medium properties were obtained from book values. The wants to protect the uterus. transducer model was based on the specifications of the Sonablate 500 HIFU probe. The distribution of the seeds was extracted from patients’ medical images. The target lesion volume, location and order of sonica- P66 tions were selected based on standard treatment protocols. The ab- Patient selection by screening pelvic MRI in MR-guided focused lated lesion volume was estimated from the model data using a ultrasound surgery of uterine fibroid thermal dose metric based on cumulative equivalent minutes. 1 2 1 1 1 Junhai Zhang , Xiaoxia Liu , Hairui Xiong , Zhenwei Yao , Qian Zhou , RESULTS The simulation results show that the EBRT marker obstructs 1 1 Haoxiong Li , Ying Tang the propagation of the ultrasound beam and distorts the focal Department of Radiology, Huashan Hospital, Fudan University, volume when positioned within 5mm of the focus (Fig. 1). The distor- Shanghai, China; Department of Gynecology, Obstetrics & Gynecology tion significantly increases when a large number of seeds is intro- Hospital, Fudan University, Shanghai, China duced in the model. Both the seeds and the markers act as Correspondence: Junhai Zhang scatterers, reflecting energy away from the intended focus. This leads Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P66 to two undesired implications. Firstly, the intended treatment region Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 94 of 122 receives less energy and as a result thetissue is not adequately thrombosis; three patients reported lower extremities numbness, two heated. Secondly, the wave reflected by the seeds induces heat- patients had soft tissue edema around the lesions. ing at unintended regions, which often include healthy tissue CONCLUSIONS MRgFUS is effective for short-term pain palliation of and the rectal wall. bone metastases. Such noninvasive technique is safe and can im- CONCLUSIONS The distortion introduced due to the presence of the prove patients’ living condition. EBRT marker, and more importantly, due to the presence of the brachy- therapy seeds may result in undertreated regions due to less energy ar- riving at the focus or overtreated regions due to reflections which may affect organs at risk. Depending on the position of the targeted regions P69 and the distribution of the markers or seeds, both effects may be un- Characterization and validation of a 64-element Capacitive Micro- desirable. The results presented have particular importance for patient Machined Ultrasound Transducer (CMUT) annular array with a 256 selection and treatment planning for prostate salvage-HIFU after failed element imaging array in the same plane, capable of tissue EBRT or brachytherapy and indicate the necessity to reconsider the ablation of the prostate 1, 2 1, 2 1 treatment parameters used during theseprocedures. Christopher Bawiec , William Apoutou N'Djin , Guillaume Bouchoux , 3 4 1, 2 Nicolas Sénégond , Nicolas Guillen , Jean-Yves Chapelon 1 2 Inserm, U1032, LabTau, Lyon, France; Université Lyon 1, Lyon, Rhone- 3 4 Alpes, France; Vermon, Tours, France; Edap TMS, Vaulx-en-Velin, France Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P69 OBJECTIVES The goal of this work was the validation and characterization of a recently developed CMUT probe that is capable of delivering a therapeutic ultrasounddose while simultaneously en- abling the delivery to be monitored through ultrasound imaging. The primary application of this probe is the treatment of prostate cancer forthermally ablating the cancerous tissues. The currently used probe Fig. 1 (abstract P67). See text for description is a high intensity focused ultrasound (HIFU) transducer utilized treat- ment of prostate cancer. Thesedevices are fabricated using spheric- ally focused piezoelectric materials, however, there are drawbacks to the current design. In this work, CMUT probes were investigatedfor P68 the HIFU applications as they offer potential advantages over the The safety and short-term efficacy of MR guided focused current designs allowing decreased fabrication costs at an industrial ultrasound surgery for bone metastases-induced pain palliation scale and ease ofminiaturization. 1 1 1 1 1 Hairui Xiong , Qian Zhou , Junhai Zhang , Haoxiong Li , Ye Chen , METHODS A planar ultrasound probe consisting of a 64-element CMUT 2 1 1 1 Qiong Li , Ying Tang , Zhenwei Yao , Xiaoyuan Feng annular array around a linear 256-element CMUT imaging array was Department of Radiology, Huashan Hospital, Fudan University, studied in simulation and experimentally (impedancemetry, microscopy Shanghai, China; Department of Anesthesiology, Huashan Hospital, and vibrometry as well as hydrophone measurements and in vitro le- Fudan University, Shanghai, China sion formation). Numerical models of the therapeutic acoustic field Correspondence: Hairui Xiong were performed using the Rayleigh integral to determine the feasibility Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P68 of electronic focusing and inducing thermal lesions. For comparison, simulations of a geometrically focused existing piezo-based technology OBJECTIVES To discuss the safety and short-term efficacy of MR- (16-element annular array) was also performed. Experimentally, key guided Focused Ultrasound Surgery (MRgFUS) for pain palliation of static parameters (collapse-, snapback-, andbreakdown voltages) for bone metastasis patients. use in the different modes of operation (conventional, collapse, and METHODS 14 patients with painful bone metastases were recruited collapse-snapback) were identified. The HIFU capabilities of the device in a prospective study. The treating efficacy is characterized by Nu- were also investigated experimentally (pressure field and radiation merical Rating Scale (NRS), theBrief Pain Inventory Quality of Life force measurements) with the creation of in vitro lesions. The imaging (BPI-QOL) survey, and Karnosky Performance Status Scale (KPS). the capabilities were also compared between the existing device and the adverse events occurred pre and post-treatment were analyzed. Nor- prototype in order to validate the modelling. mal distributed statistics were analyzed with paired-samples t test, or RESULTS Simulations showed that the planar CMUT design could Wilcoxon rank sum test was used. dynamically focus from 3-7cm and create lesions comparable in RESULTS 14 patients were treated with MRgFUS. two patient size and shape to the geometrically focused transducer. Microscopy dropped out of the study. The NRS ratings are 6.5(4), 5(5.25), 2.5(5), allowed visualization of the static collapse and snapback of individ- 2.5(4.75) for pre-treatment, one week, one month, two months, and ual cells which was confirmed in parallel with impedance measure- three months, respectively. Such variances of NRS ratings are statisti- ments. Collapse occurred at 132.5±5V and snapback occurred at 95 cally significant (P<0.05). The BPI-QOL ratings are 42.42±8.27,30.67 ±5V. Dynamic behavior of the CMUT (vibrometry) in air and in ±12.29, 29.17±15.38, 29.92±17.67, 35.67±19.28, respectively. The BPI- water allowed identification of the collapse-snapback operational QOL ratings decrease in the first two months after the treatment mode. In this mode, the probe was capable of generating up to which is statistically significant (P<0.05) ; whereas for the third 5W/cm2 surface intensity. Furthermore, experimental validation month, the BPI-QOL rating is statistically insignificant compared with performed on the probe utilizing hydrophones and in vitro tissue the one before the treatment. The KPS ratings are80(28), 80(20), confirmed that the probe was capable of creating lesions in tissue. 65(45) for pre-treatment, one week and three months, respectively. CONCLUSIONS The results of the investigation confirmed that the Three months after the treatment, the KPS ratings decreases which is current prototype is capable of creating lesions in biological statistically significant compared with the one before the treatment tissue. Both the simulations and the experiments were able to (P<0.05). After the treatment, one patient developed deep venous confirm the capability of this ultrasound probe. The ultrasound Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 95 of 122 imaging part of the probe also provided an enhancement to the experiment conditions. The system with the software light field can spatial resolution of the existing technology which allowed for a measure the average intensity of sonoluminescence. higher resolution improving the localization of the treatments. RESULTS The intensity of multibubble sonoluminescence was in- This new ultrasound probe prototype has the potential of provid- creased, when ambient pressure increased from 0.1MPa to 10MPa. ing an improved treatment method of prostate cancer that can However, the region of cavitation cluster decreased gradually. increase the quality of the treatment and subsequent outcome of CONCLUSIONS The increasing ambient pressure caused active the patient while adding the benefits of a reduction in the cost area of cavitation cluster to decrease. Besides the cavitation clus- and size of the probe. This project was supported by the French ter activity will be more drastic, when electronic power reaches a Single Interministerial Fund (FUI, 2013). certain value. P70 P72 Precision microvascular therapy via the synergy of light and sound 1,2 2,3 2,3 3 3 4 5 Method of temperature estimation in skin for a focused ultrasound H. Zhang ,Z. Hu ,J.Li , Q. Liu , S. Yuan , Y. Paulus , X. Yang , 1,2 applicator using a multi-validated numerical thermal model and X. Wang Aghuinyue E. Umenei, Ziqi Wu Institute of Acoustics, Tongji University, Shanghai, P.R. China; Global Discovery, Amway, Grand Rapids, Michigan, United States Department of Biomedical Engineering, University of Michigan, Ann Correspondence: Aghuinyue E. Umenei Arbor, MI, USA; Department of Ophthalmology, the First Affiliated Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P72 Hospital of Nanjing Medical University, Nanjing, P.R. China; Department of Ophthalmology and Visual Sciences, University of Michigan, Ann OBJECTIVES Focused ultrasound through ablation has gained increasing Arbor, MI, USA; Department of Mechanical Engineering, University of use in aesthetic applications over the last decade. Recent studies show Kansas, Lawrence, KS, USA non-ablative focused ultrasound also provided clinical benefits through Correspondence: H. Zhang localized heating. Temperature measurements are critical for efficacy and Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P70 safety for such applications but are not easily or cheaply obtainable under in vivo dermatological conditions. Current temperature measure- Angiogenesis and neovascularization are hallmarks for a variety ment methods are expensive, time consuming, application specific/organ of pathological conditions, including cancer and many eye dis- dependent and provide unsatisfactory spatiotemporal resolution. We eases, and play a crucial role in disease onset and progression. propose a method including a multiphysics model to quickly estimate Antivascular therapies that aim at either removing microvessels skin temperature distribution in 2D, which was validated using multiple or slowing down their growth represent a proven new strategy approaches and lends itself to an iterative development process. to intervene the progress of these conditions and improve the METHODS An acousto-thermal 2D FEA multiphysics model was used prognosis. Here we report the development of a photo-mediated to estimate focused ultrasound thermal distribution in a tissue mimic ultrasound therapy (PUT) technique as a new concept of localized materials (TMMs) which are acoustically similar to skin tissue, by in- antivascular therapy. Unlike any of the previous optical- or corporating acoustic field 2D surface measurements of the 3W appli- ultrasonic-alone techniques, laser pulses and ultrasound bursts cator from its hydrophone scans (Fig. 1). The effects of ultrasound on are synergistically applied in PUT, which facilitate noninvasive tissue heating were modeled in two phases with parameters both treatment of subsurface microvessels with unprecedented high measured in the lab and obtained from literature where possible. Ini- precision. PUT takes advantages from the high native optical con- tially, the acoustic pressure in theTMM generated by the applicator trast among biological tissues, and has the unique capability to was initially solved using wave equations in the frequency domain. self-target on blood vessels without causing unwanted damage The thermal distribution in the tissue was subsequently calculated to surrounding tissue. As demonstrated through the experiments using the obtained acoustic energy as the heat source term to the on animal models, PUT can treat microvessels in target tissue via bioheat equation in a time domain simulation over 7 seconds (Fig. 2). different mechanisms, including blocking local vessels by indu- The simulated transversal 2D temperature distribution was compared cing blood clots or disrupting vessels causing local hemorrhage, to thermocouple measurements in the tofu for multiple applicators each with values in clinic. Moreover, PUT working at different (10). The thermocouple measurements were taken at 0.5mm spacing optical wavelengths can selectively treat veins or arteries by along the longitudinal axis of the applicator, at 2, 4, 6, and 8 mm below utilizing the contrast in optical spectra between deoxy- and oxy- the surface to give a 2D measurement grid of the TMM. The focal dis- hemoglobin. PUT, as a novel antivascular therapy technique with tance of the ultrasound applicators were measured to be 4±1 mm in the capability to precisely target vessels and precisely control the water. Furthermore, simulations were performed using a porcine treatment effects, holds promise to impact clinical management muscle model and the results were compared in a similar fashion to of cancer and eye diseases by delivering optimized treatment MR thermometry acquired in muscle samples. The validated multiphy- outcome with minimized complication. sics model was finally used to estimate a thermal distribution for multi- layered skin model. RESULTS The model estimated average maximum temperature rise from nominal (37 C) in tofu at the focus, at 6.9 C compared to 6.56 C from the P71 thermocouple measurement along the focal line of interest. A good cor- Effect of ambient pressure on cavitation bubbles at the focal point relation (R2 = 0.89) was seen between modeling results and thermo- of a spherical resonator with open ends couple measurements in tofu for 10 ultrasound transducers (Fig. 3). For 1,2 1, 2 1 1, 2 Zonggui Chen , Huan Liu , Xiaobo Gong , Faqi Li porcine muscle, model estimated temperature rise along focal line of State Key Laboratory of Ultrasound Engineering in Medicine Co- interest of 14.9 C whereas MR thermometry measured 14 C. Model accur- Founded by Chongqing and the Ministry of Science and Technology, acy was between 88%-93% once equipment error was factored, provid- Chongqing, China; College ofBiomedical Engineering Chongqing ing an acceptable temperature estimate for developmental research at Medical University, Chongqing, China the high resolutions need for skin tissue. Skin multilayer model estimated Correspondence: Zonggui Chen the maximum temperature was 19 C at the focal region. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P71 CONCLUSIONS Model estimated temperature in tissue mimic mater- ial and porcine muscle were validated with thermocouple and MR OBJECTIVES Study on the activity of cavitation cluster under the differ- thermometry approaches. Byinputting free-field acoustic field infor- ent ambient pressure by the average intensity of sonoluminescence. mation and acousto-thermal properties of tissue media and building METHODS The experiments, activities of cavitation bubbles and the a validated 2D model, we have the proposed an easy method of esti- sonoluminescence are captured by general camera and emICCD mating temperature changes in multilayered tissue such as skin with under different ambient pressure, are conducted in the same close approximation at a low cost and good accuracy. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 96 of 122 P73 A 3D-printed skull model with corresponding acoustic characteristic of human skull for ultrasound brain imaging and diagnosis Chen Bai, Meiling Ji, Dui Qin, Mingxi Wan The Key Laboratory of Biomedical Information Engineering of Ministry of Education, Department of Biomedical Engineering, School of Life Science and Technology, Xi`an Jiaotong University, Xi'an, China Correspondence: Chen Bai Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P73 OBJECTIVES Recently, more and more attention has been paid to the ultrasound brain imaging as it is non-radioactive compared with CT and low-cost compared with MRI. And the existence of skull limits the propagation of the ultrasound and then limits the frequency of transducer to be transmitted and furtherly limits the resolution ofthe resulting image. To realize the noninvasive transcranial brain im- aging, the foreknowledge of the acoustic characteristic of the skull is essential. The ossa temporale (temporal bone), which is the thinnest part, has been proved that is the best acoustic window for imaging. However, the intact human skull was precious and hard to get the part we just need by excision. Instead, with the help of the 3D print- ing technology and related software, this problem can be overcome easily. Furthermore, various invitro research experiments about brain Fig. 1 (abstract P72). MRI image of applicator in pork muscle (axial imaging can be proceeded with the 3D-printed model which section) after 7s showing thermal profile, and calculated temperature rise matches the acoustic characteristic of human skull. METHODS The acoustic attenuation coefficient and ultrasonic sound velocity for the temporal bone was firstly measured. The human skull was mounted on a threedimensional scaffold and immersed in a tank with degassed water. Two single element transducers were placed on the walls of the tank coaxially on each side of the skull manually parallel to the temporal bone, one for transmitting signal and the other for receiving. The transducer was excited by a 5-cycle Sine sig- nal whose peak to peakvalue was 2 V and the receiving signal was collected with the NI data acquisition system. The direct ultrasonic wave in free field was normalized and recorded as a reference signal for calculating the attenuation coefficients and velocity. Each time, we measured at 10 different points of the temporal bone, and each point was repeated ten times and get the mean value (Fig. 1). For ensuring the optimum frequency for transcranial experiment, 5 pairs of transducers with frequency ranging from 1 to 5 MHz were tested. Moreover, the CT scan images of human skull was obtained, with 1mm interval between 2 layers, to measure the thickness of bone, and furtherly, to design and modify the model by MIMICS for 3D print. After that, to find a certain material matching the acoustic characteristics of the human skull and the cerebral vessels, the at- tenuation coefficients and velocities of 9 different kinds of materials including resin, nylon, PEEK, PLA, etc. with different thickness were measured. Additionally, considering the effect of the temporal bone’s curvature, the materials were modeled in the shape of the temporal Fig. 2 (abstract P72). Simulated thermal model of TMM (tofu), bone rather than flat plate. By aid of the software MIMICS, we got the showing ultrasound applicator and temperature distribution after satisfying models with excision of the redundant part of the skull. 7s (longitudinal section) RESULTS According to the results of measurement, 1.8 to 2 MHz was determined with synthetic consideration of the attenuation, propagation and the resolution for further imaging. Hence, the averaged attenuation coefficient and velocity of the temporal bone for the human skull model measured with 1.9 MHz frequency were 7.34 dB/mm and 2398.13 m/s, respectively. And for the optimum material within the 9 kinds of mate- rials, PLA, those were 7.11 dB/mm and 2270.24 m/s under the same test- ing condition, which means the thickness of the 3D-printed skull model with the PLA should be the same as that of human skull. In other words, the mean thickness of the temporal bone of the printed skull model using PLA was about 1.8 mm, in correspondence with the acoustic char- acteristic of human skull with the mean thickness of1.7 mm. For another, the material to print the vascular model, one kind of resin, the SPR6000B, with averaged attenuation coefficient 0.8 dB/mm and velocity 1963 m/s, approximating to the results in the free field, was picked out. Finally, by Fig. 3 (abstract P72). Surface map and line measurement of measuring the performance of the new models, it has been proved that temperature rise of ultrasound applicator in tofu after 7s the human skull and cerebral vessels can be conveniently and com- (longitudinal section) pletely replaced by the 3D-printed models. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 97 of 122 CONCLUSIONS In order to design a skull model which can completely transducer with periodic array surface was designed to realize replace the intact human skull in ultrasound brain imaging study, we focusing enhancement. measured the acoustic characteristics of temporal bone of human skull, METHODS To verify the enhanced acoustic focusing of concave and moreover, a number of different materials to ascertain the material transducer with periodically aligned subwavelength grooves, the whose acoustic attenuation characteristic was similar to the human acoustic pressure and temperature elevation generated by the new skull or the vascular. The results showed that PLA met the requirement structure were investigated based on both experimental measure- for skull model and the attenuation coefficient of SPR6000B was small ments and numerical simulations, and then compared with the re- enough to print the vascular model. Finally, we have acquired the satis- sults obtained from conventional concave transducer. The software fying skull model with excision of the cranial skullcap, and to fix the de- of Comsol Multiphysics was employed to perform the numerical sim- signed vascular model conveniently. In the following study of brain ulations based on the combination of2D Helmholtz equation and imaging, a scheme that has the contrast agents circulate in the vascular bio-heat transfer equation. An optical fiber hydrophone and thermo- model fixed on the skull model, and manage to imagethe vascular couple were utilized to measure the acoustic field and detect the through the temporal bone via a 128-element linear array transducer temperature rise at the focus in a tissue phantom exposed to HIFU with 2 MHz frequency will be performed. pulses, respectively. RESULTS Both the experimental measurements and numerical simu- lations show that the peak pressure amplitude and the rate of temperature rise at the focal region are obviously larger in the case of corrugated transducer rather than the conventional transducer as Fig. 1. Different from previous studies originated from planar arrays of periodic structure, the application of spherically curved arrays of periodic grooves could result in extraordinary acoustic transmission close to Wood’s anomaly that at the wavelength slightly less than the groove periodicity. CONCLUSIONS Numerical simulations and experimental measure- ments both show that, comparing with conventional concave trans- ducer, the concave transducer with periodic array of subwavelength grooves is more efficient to improve acoustic focusing and enhance relative bioeffects. This work may open new possibility to design more favorable focused ultrasonic transducers that could significantly improve HIFU treatment effects. Fig. 1 (abstract P73). Receiving signals in the (a) free field, (b) skull, (c) PLA and (d) SPR6000B; (e) the CT scan image and (f) the 3Dprinted model Fig. 1 (abstract P74). Simulated acoustic pressure lever and temperature distributions on the focal plane: acoustic pressure lever with (a) the corrugated transducer and P74 (b)conventional transducer; and temperature distribution in Enhanced ultrasonic focusing and temperature rise by using phantom exposed with (c) the corrugated transducer and (d) sub-wavelength periodic structure conventional transducer Chenghai Li, Xiasheng Guo, Juan Tu, Dong Zhang, Zhou Lin Key Laboratory of Modern Acoustics (Nanjing University), Ministry of Education, Nanjing University, Nanjing, Jiangsu, China Correspondence: Chenchen Bing Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P74 P75 Optimization of thermal fields by focal region modulation in HIFU OBJECTIVES High-intensity focused ultrasound (HIFU) has been brain tumour treatment used in clinic as a non-invasive therapeutic method to treat solid 1 2 1 1 Shihui Chang , RUI CAO , Yabin Zhang , Shijing Wu1, Xiqi Jian tumors for decades. Although various (e.g., lens, spherically 1 2 Tianjin Medical Univercity, Tianjin, China; Tianjin University of Science & curved transducer and multi-element phased array) were devel- Technology, Tianjin, China oped to focus ultrasonic energy, the ultrasonic focusing efficiency Correspondence: Shihui Chang is highly restricted by the size of transducers due to the limita- Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P75 tion of manufacture technology. In the present work, a concave Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 98 of 122 OBJECTIVES The temperature rise during HIFU brain tumor treatment convective cooling and acoustic streaming can change the should be controlled accurately to avoid side effects by overheating temperature considerably near blood vessel. such as cerebral hemorrhage. In this work a modulation method to CONCLUSIONS Three dimensional simulations of focused ultrasound control the shape and volume of focal region as well as the ablation has been perfomed in a patient specific geometry. Simula- temperature distribution at focus was performed by adjusting the tion using multiple GPUs can sufficiently reduce the simulation time driving signals of the transducer elements. and can help to construct surgical planning platform. High ultra- METHODS The simulation model was reconstructed based on the CT sound power and nonlinear propagation effects with appropriate data of a volunteer's head and the 82-element transducer. In this treatment planning can sufficiently reduce the treatment time. We study, the finite difference time domain (FDTD) method was used to theoretically showed that the treatment time can be reduced to few calculate the temperature distribution induced by HIFU in the brain. minutes. The presented model can be used in planning tools for the Two foci with a certain distance were generated on the acoustic axis thermal ablation of tumour in other organs. by a HIFU source using two driving signals which were determined by time reversal method. This double foci fusion method was applied to create a uniform temperature distribution at focal region and ad- P77 just the length and volume of the focal region. Principal component analysis of acoustic aberrations for rapid RESULTS Thermal field distribution at focal region can be ad- HIFU beam refocusing: a simulation study justed by changing the time delay of the two driving signals. Xiaoxu Lei, Martijn de Greef, Chrit Moonen, Mario Ries Thus, a uniform distributed temperature field can be created at Imaging, University Medical Center Utrecht, Utrecht, Netherlands focal region. What is more, this method was able to adjust the Correspondence: Xiaoxu Lei length andvolumeofthe focalregionusing thesametotal Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P77 acoustic energy by varying the distance between the two foci. CONCLUSIONS The HIFU temperature distribution, shape and volume OBJECTIVES The propagation path of the HIFU beam frequently in- of the focal region could be modulated by the double foci fusion troduces acoustic aberrations, which degrade the focus quality. Non- method. The focal region volume formed by a single irradiation could invasive refocusing 1-3 based on acoustic radiation force imaging be enlarged effectively while avoiding the brain tissue overheated. This (ARFI) have been proposed, which address this problem. However, method can provide basis for the focal temperature distribution modu- the refocusing process is generally lengthy and cumbersome, due to lation in the treatment of brain tumor in further study. the multitude of independent measurements. Here, we evaluate the feasibility of accelerating this process by replacing the Hadamard (H) base and Zernike polynomials (ZP) base of the original approaches P76 by a new base which is derived using principal component analysis Modelling and simulation for the focused ultrasound ablation of (PCA) on simulated phase aberrations from a uterine fibroid model. liver tumour METHODS The pressure at HIFU focus can be expressed as the sum 1,2 2 1 1 Maxim Solovchuk , Tony W.H Sheu , Manuel Diaz , Peter Deng of the product of the acoustic wave fired from each individual trans- Institutes of Biomedical Engineering and Nanomedicine, National ducer element and the acoustic aberration experienced along its 2 N Health Research Institutes; Engineering Science and Ocean Engineering beam path:p = ∑ =1 gi.χi, (1)where χi is the complex source term of Department, National Taiwan University each emitter, gi the aggregated attenuation and phase shift along Correspondence: Maxim Solovchuk the ith propagation path for the element and p the complex focus Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P76 pressure of all elements.For non-iterative adaptive refocusing approach, Eq.1 can be transformed into matrix form:pX = g.X, OBJECTIVES High intensity focused ultrasound is very promising new (2)Where pX is a 1 x K complex row vector whose elements are the technology, that has many therapeutic application, among them are focus pressures corresponding to each excitation pattern from base the treatment of cancer indifferent organs. One of the main diffi- X, g is a 1 x N complex row vector containing the aberration cultlies, that limits further development of HIFU, is very difficult treat- information for each transducer element, X is the N x K orthogonal ment planning and unpredictable behavior of the necrosed area in base whose column vectors are used for the actual measurement the presence of bubbles. Computational fluid dynamics can greatly process, which requires 4xK independent measurements. Both, a help in this regard. Calculation of ultrasound propagation in a patient Hadamard base H and Zernike polynomials base Z have been specific geometry is very time consuming process. In order to applied as X in Eq.2 for refocusing the beam. In order to accelerate achieve reasonable simulation time HPC (high performance comput- the refocusing process, the less numbers of columns of X used to get ing) should be used. The problem will be solved using HPC on mul- accurate g in Eq.2 the better. Here, a method is proposed to apply tiple GPUs. This work is a step towards the development of surgical PCA on prior knowledge of the phase aberrations obtained from planning platform for a non-invasive HIFU tumour ablative therapy in acoustic simulations to form an orthogonal base (PCA base), which a real liver geometry. allows to describe the dominating phase aberrations in a more METHODS The computational model has been developed for the compact form and thus allows to significantly reduce the prediction of temperature elevation in the patient specific liver measurement process. Simulation studies were conducted based on geometry. The developed computational model is based on the non- 4 MRI datasets of uterine fibroid patients, which were obtained linear Westervelt equation with relaxation effects being taken into during MR-guided HIFU therapy. For each patient, 343 different focus account, bioheat equations for the perfused tissue and blood flow positions within the ablation volume were chosen and the domains. The nonlinear Navier-Stokes equations are employed to de- corresponding phase aberrations for each focus position were obtained scribe the flow in the large blood vessels. A new equation is derived using a stochastic ray tracing method4. Subsequently, 342 positions for the description of large amplitude oscillations of bubbles in visco- served to obtain the new compact base, which was finally validated for elastic medium. The cavitation model is coupled with acoustic and rapid autofocussing. thermal models. RESULTS As Fig. 1 shows, PCA based autofocussing typically re-estab- RESULTS Three dimensional field coupling computational study has lished 90% of the fully phase corrected focus pressure after only 16 been performed. Explicit symplectic finite difference scheme has measurements, whileZernike and in particular Hadamard based cor- been developed for the solution of full wave equation. We used mul- rection require a significantly (5-20 times) longer measurement time, tiple GPUs for the modeling of nonlinear wave equation. to achieve the same improvement. Figure 2 compares the pressures Temperature elevation by focused ultrasound in a minipig has been maps when applying different numbers of PCA base modes and indi- studied experimentally by MRI and numerically. The method for the cates that using about 10 modes are able to reduce the side lobes non-invasive thermal tissue properties has been proposed using MRI. and have similar results as fully phase compensated. Good agreement between the predicted and measured results has CONCLUSIONS Although ARFI based HIFU autofocusing for the com- been obtained. It was also shown that in large blood vessel both pensation of beam aberrations has shown a substantial promise, so Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 99 of 122 far the approach is frequently limited by unacceptably long measure- P78 ment time of several minutes per focus point. Here, we demonstrate A fast 3-D transcranial focused ultrasound simulation based on ray that autofocusing using an individually adapted base derived from tracing 1, 2 2 1,2 prior knowledge can substantially shorten this process. Changzhu Jin , John Snell , Dong-Guk Paeng Ocean System Engineering, Jeju National University, Jeju, Korea; References Focused Ultrasound Foundation, Charlottesville, Virginia, USA [1] Herbert et al., IEEE Trans. Ultrason. Ferroelectr. Freq. Control 56(11) 2388 – 99, Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P78 [2] Larrat et al., IEEE Trans. Ultrason. Ferroelectr. Freq. Control 57(8) 1734 – 7, OBJECTIVES Two objectives are explored: improved skull aberration 2010 correction and focal spot quality visualization. Others have used full [3] Kaye et al., Medical Physics 39 6254 (2012) wave acoustic simulation to explore the possibilities of more accurate [4] Koskela et al., J.Acoust.Soc.Am. 136(3), 2014 phase correction and the simulation and visualization of aspects of focal spot quality [1]. However, the computational cost of such complete acoustic simulations prevents, at the current time, their routine clinical use. In current clinical practice, the location and qual- ity of the hotspot must be evaluated and corrected for during the procedure itself through the use of a number of low-power sonica- tions prior to a final therapeutic sonication [1]. It would bebeneficial to have pre-surgical planning information to inform patient selection and procedure planning in advance of the delivery of energy to a pa- tient. Patient-specific skull characteristics and particular anatomical targets within the skull can make achieving prescribed focal spot temperature elevations difficult or impossible. The focal spot quality in terms of location and temperature could be improved by chan- ging transducer parameters, relative angle between the transducer and patient skull, and morerealistic phase correction. A phase correc- tion taking account of realistic parameters with sufficiently fast com- putation is desired to accomplish these optimizations closer to or within the clinical workflow. In this work, a simplified acoustic simula- tion tool based on GPU-accelerated ray tracing was developed to cal- culate skull aberration correction and to provide real-time visualization of an estimated three-dimensional (3D) pressure field around the targeted focal spot. METHODS A modular 3D planning software system was developed Fig. 1 (abstract P77). The focus pressure versus number of base for transcranial focused ultrasound procedure planning. Within this mode used curves for PCA base, ZP base and H base. Note the rapid environment, CT and MRimages can be displayed and registered. convergence of the PCA acceleratedmeasurement process, which Skull surfaces in the CT image are characterized using a threshold- only takes 4 base modes while ZP and H based process requires based algorithm and a 3D edge detection operator. A single layer much more modes skull model was assumed. A virtual FUS transducer which has a 30cm diameter hemispherical shape and consists of 1024 Fibonacci pat- terned elements was designed. Each transducer element acoustic beam is represented by a computed path through the cooling water, skull and brain tissue. This path from the transducer to the focal point was divided into 3 individual ray segments (water, bone, and brain). The refraction caused by impedance difference at skull bone surfaces was computed based on Snell’s Law. Any ray with an inci- dent angle at the outer skull surface greater than critical angle was neglected. The computation volume of pressure field contains 21×21×21 cubic sample points and the distance between each point on XYZ axis is 1mm. The center of the computation volume was set to lie on the target point. The projection vector of each sample point onto the third ray segment was computed and the length projection was taken as new third ray segment. In order to get a focused spot, the phase at the end of the ray trace was collected and utilized for phase correction prior to pressure computation. The pressure field contribu- tion from each ray trace to the sample point was accumulated using a simplified beam profile model. The attenuation coefficient of water, skull, and brain was set as 0.0022, 0.3103 and 0.0690 dB/cm/MHz re- spectively. The density of the water, skull, and brain was set as 1000, Fig. 2 (abstract P77). Simulated pressure maps obtained after 1x4, 1900 and 1030kg/m^3. The assumption of exponential decay on the 5x4, 10x4, and 256x4 measurements of the autofocusing process. pressure along longitudinal direction was applied and a sinc function Note the pressure map acquired after 10x4 measurements is largely was implemented as pressure decay on the axial direction of the beam. the same as fully phase compensated one, indicating a rapid RESULTS This tool provides an intuitive 3-D view of pre-treatment refocusing process imaging data and also improves the understanding of the focal Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 100 of 122 OBJECTIVES To investigatethedifferentvasculareffectofrabbitVX2 tumor induced by diagnostic ultrasound combined with microbubbles (MB). METHODS Forty rabbit with VX2 tumor were treated by diagnostic ultra- sound with different strengths of Mechanical index (MI) 0.3,0.7,1.4 with MB as experiment group (n=10, n=10, n=10) and negative control (n=10) for 5 minutes. 0.1ml microbubbles in 5 ml normal saline was used as a therapeutic dose. The tumor blood perfusion was imaged with contrast- enhanced US (CEUS) before and after treatment which were analyzed by the contrast analysis software of VINNO 70 . After the treatment, the tumor samples were collected for pathological examination. RESULTS The PI in MI=0.3 increased after treatment, the PI and Fig. 1 (abstract P78). (a) The visualization of wave beam path AUC of MI=1.4 decreased after treatment (Fig. 1). using ray tracing method. (b) The Sagittal plane of simulated CONCLUSIONS Low-intensity diagnostic ultrasound (MI=0.3) com- focal pressure bined with microbubbles can enrich the blood perfusion of VX2 tumor in rabbits, and high-intensity diagnostic ultrasound (MI=1.4) can decrease the blood perfusion. quality on various FUS transducer setups interactively. The pressure field which takes account of attenuation and transmission loss was calculated and displayed within 1 Hz computation speed (Fig. 1b). CONCLUSIONS A fast 3D patient data visualization and acoustic simu- lation software based on ray tracing method were developed (Fig. 1a). The phase correction andtransducer setup could be exported to a third party full-wave simulation software for more accurate prediction of the pressure field. It offers help to the clinician with patient selection and to decide the optimum transducer setup during pretreatment planning. Validation of the accuracy of the pressure field com- puted by this method as compared with full wave simulation methods and hydrophone data is the subject of future work. Reference [1] Clement, G.T. and K. Hynynen, Phys Med Biol, 2002. 47(8): p. 1219-36. P79 Efficacy of ultrasound mediated microbubbles in diclofenac gel to Fig. 1 (abstract P80). Ultrasound contrast image of VINNO70 before enhance transdermal permeation in rheumatoid arthritis induced and after the treatment.(A, B, C, D on behalf of MI=0.3,0.7,1.4 group rat 1 2 and control group, 1, 2 and 3 respectively represent the twodimensional Ai-Ho Liao , Ho-Chiao Chuang ultrasound,ultrasound contrast image before treatment and after National Taiwan University of Science and Technology, Taipei, Taiwan; treatment) Images show that thegroup of MI=0.3combined with National Taipei University of Technology, Taipei, Taiwan microbubbles can enrich the blood perfusion of VX2 tumor in rabbits, Correspondence: Ai-Ho Liao MI=1.4 group decrease the blood perfusion,and there was no significant Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P79 change in the blood perfusion of the MI=0.7group and control group before and after treatment This abstract is not included as it has already been published: Liao A.H, Chuang H.C, Chung H.Y. Efficacy of ultrasound mediated microbubbles in diclofenac gel to enhance transdermal permeation in rheumatoid arthritis induced rat. IEEE. 2015. Available from: P81 http://ieeexplore.ieee.org/document/7319152/. Augmentation of muscle perfusion by microbubble mediated ultrasonic cavitation W. Gao, C. Yi, X. Qiao, Z. Liu Department of Ultrasound, Xinqiao Hospital, Third Military Mediical University, Chongqing, China P80 Correspondence: W. Gao Vascular effect of rabbit VX2 tumour induced by diagnostic Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P81 ultrasound with microbubbles Xueyan Qiao, Zhong Chen, Cuo Yi, Wenhong Gao, Shunji Gao, OBJECTIVES Ultrasound has been known capable of enhancing Zheng Liu tissue perfusion through both thermal and non- thermal bio- Ultrasound Department, Xinqiao Hospital, Third Military Medical effects. The main purpose of this study is to investigate if the en- University, Chongqing, China hanced non-thermal ultrasonic bioeffect- microbubble mediated Correspondence: Xueyan Qiao cavitation could effectively agitate blood perfusion of mice skel- Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P80 etal muscle. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 101 of 122 METHODS Twelve healthy nude rice were intravenously injected microbubbles for both ultrasonic cavitation and contrast imaging. Firstly, one randomly chose posterior limb underwent microbubble mediated cavitation, while the other side went through no treatment as an own control. Intermittent ultrasound applied for cavitation was set at a frequency of 4 MHz, pulse length of 18 cycle, pulse repetition frequency of 50 Hz, mechanical index of 1.4, work/rest interval of 0.48s/2s, and total irradiating time of 10 minutes. Microbubbles for cavitation with a number of approximately 8x10 was injected slowly meanwhile irritate ultrasound was applied. Then simultaneous contrast imaging of both limbs through a transverse section was conducted right after cavitation procedure (Fig. 1). Dynamic contrast video was recorded for evaluating of skeletal muscle perfusion, and quantitative parameters including peak intensity, area under curve, and ascending slope were analyzed. Average values of treatment side and control side were statistically compared using paired sample T test. RESULTS Contrast imaging records exhibited prominent acceleration and abundance of treated muscle perfusion. Entry of microbubbles into peripheral circulation is much more rapid and numerous in the treated side than in the control side (Fig. 2). Quantitative analysis re- sults also showed a steady rise of perfusion on the treated side with statistical significance. Average values of peak intensity, area under curve and ascending slope in control side were (71.2467±10.86445), (6436.7658±981.94242), and (0.7342±0.30417), while in treated side they were (82.0100±11.23804), (7584.8033±916.03143), and (0.9208 ±0.39867),respectively (P<0.05) (Fig. 3). CONCLUSIONS Microbubble cavitation irritated by high mechanical, low duty cycle intermittent ultrasound could effectively agitate mice skeletal muscle perfusion, leading significant raise of both blood vel- ocity and blood volume. Fig. 1 (abstract P81). Illustration of simultaneous contrast imaging of both limbs through a transverse section Fig. 3 (abstract P81). Histogram of peak intensity value, area under curve and ascending slope in the two groups P82 Relationship between different molecule sizes and delivery efficiency of pancreatic cancer cells mediated by different cavitation dose 2,1 1 2 2 3 Lizhou Lin , Mouwen Cheng , Fan Li , Lianfang Du , Alfred C. Yu , Peng Qin Department of Instrument Science and Engineering, Shanghai Jiao Tong University, Shanghai, China; Department of Ultrasound, Shanghai First People’sHospital, Shanghai Jiao Tong University, Shanghai, China, Fig. 2 (abstract P81). Continuous screen shots of contrast imaging 3Department of Electrical and Computer Engineering, The University of after treatment procedure. The treated side (left) showed much waterloo, Waterloo, Alberta, Canada more rapid and abundant perfusion of peripheral circulation than Correspondence: Lizhou Lin the control side (right) Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P82 Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 102 of 122 OBJECTIVES Present studies have validated acoustic cavitation, trig- gered by ultrasound and microbubbles, could enhance the membrane permeability of pancreatictumor cells for improving macromolecular drug delivery. However, the efficiency of different molecular sizes inter- nalized into the pancreatic tumor after acoustic cavitation is not clear. This work aims to address the relationship between different molecule sizes and delivery efficiency of pancreatic cancer cells under different typesand dosesof acousticcavitation. METHODS The pancreatic cancer Panc-1 cells mixed with 1% SonoVue microbubbles and FITC-dextran with different molecule sizes (4 kDa, 40 kDa and 500 kDa) were placed in a tissue mimicking chamber and ex- posed to 1-MHz ultrasound with 1 kHz pulse repetition frequency, 300 cycles and different peak negative pressure (0.25 MPa, 0.6 MPa, 1.25 MPa) (Fig. 1). Another 7.5-MHz focused transducer was employed to passively receive acoustic signals (Fig. 2). The intensities of the ultrahar- Fig. 2 (abstract P82). Panc-1 cells mixed with 1% SonoVue bubbles monics and broadband components were respectively measured to (black water) and water (as control, redcurve) exposed to different quantify the dose of stable and inertial cavitation, respectively. The PNP (0.25 MPa and 1.25 MPa). 1 KHz PRP and 30% duty cycle. samples were stained by Propidium Iodide 30 min after sonication, and Frequency domain of the recorded signal in a period with 0.25 MPa then were analyzed by flow cytometry to assess delivery efficiency and (2-1) and 1.25 MPa PNP (2-3). Power of untraharmonics (2-3) and cell viability. Five replicates experiments were conducted to calculate broadband components (2-4) as a function of exposure time the mean standard deviation and statistically analyze. RESULTS 1. The delivery efficiency (5.90±1.37%) for different FITC- dextran and the necrotic ratio (4.44±1.05%) were not enhanced after Panc-1 cells underwent stable cavitation, compared with the control without exposure to ultrasound (3.41±1.37% and3.08±0.51% for delivery and necrosis ratio, respectively).2. No significant difference (P > 0.05) among the delivery efficiency of FITC-dextran with differ- ent mass was observed when Panc-1 cells were exposed to different inertial cavitation dose (ICD) (Fig. 3). Relatively high ICD induced 30.63±7.73%, 27.57±6.59% and 26.11±5.62% delivery ratio for 4 kDa, Fig. 3 (abstract P82). Delivery efficiency of different FITC-dextran 40 kDa and 500 kDa FITCdextran, respectively.3. Both the delivery (3-1: without ultrasound exposure; 3-2: 4 kDa; 3-3: 40 kDa; 3-4:500 kDa) efficiency for all FITC-dextran and the necrotic cells were positively and necrosis analyzed by flow cytometry Pan-1 cells mixed with 1% correlated with ICD (delivery and necrosis ratio increased by approxi- SonoVue bubbles and FITC-dextran with different molecule size mate 40%and 60%, respectively when ICD increased about 8 times). (4 kDa, 40 kDa and 500 kDa) expsoed to high inertial cavitation dose, The number of the internalized FITC-dextran molecules exhibited corresponding to 1 MHz untrasound 1.25MPa PRP, 1 KHz PRP and negative correlation with molecule size (The fluorescence intensity of 300 cycles the internalized 4 kDa FITC-dextran was almost 110-fold high than that of 500 kDa FITC-dextran, respectively) (Fig. 4). CONCLUSIONS Stable cavitation could not enhance the internalization of macromolecule into pancreatic cancer cells, while the delivery ratio of different molecule sizes was positively correlated with the inertial cavitation dose. Although the delivery rate induced by inertial cavita- tion was not dependent on the macromolecule size below 35 nm, the number of internalized molecules was negatively correlated with the molecule size. We believe these results would provide useful informa- tion for pancreatic tumor therapy mediated by acoustic cavitation. Fig. 4 (abstract P82). Delivery efficiency (4-1), necrosis ration (4-2) and the relative fluorescence intensity of internalized FITC-dextran analysis when the Panc-1 cells mixed with 1% SonoVue bubbles and FITC-dextran with different molecule size (4 kDa, 40 kDa and 500 kDa) exposed to different intertial cavitation doses (***, p<0.001; NS p<0.05) P83 Electron microscopy of renal boiling histotripsy lesions created in an in vivo porcine model 1 2 3 1 1 Y. Wang , S. Buravkov , V. Chernikov , T. D. Khokhlova , G. R. Schade , 1 1 1, 2 W. Kreider , A. Maxwell1, M. Bailey , V. Khokhlova 1 2 University of Washington, Seattle, Washington, USA; M.V. Lomonosov Moscow State University, Moscow, Russian Federation; Research Institute of Human Morphology, Moscow, Russian Federation Correspondence: Y. Wang Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P83 Fig. 1 (abstract P82). Experimental Setup for ultrasound exposure OBJECTIVES Boiling histotripsy (BH) uses millisecond-long pulses of and cavitation detection focused ultrasound (FUS) waves with shocks to mechanically Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 103 of 122 homogenize tissue. Histological evaluation has demonstrated that le- P84 sions with microscopically fine borders can be generated. Here we Subcellular localization of sonosensitizer for autophagy report a preliminary evaluation of the ultrastructural characteristics of cooperated apoptosis in sonodynamic therapy renal BH lesions created in an in vivo porcine model. L. Song METHODS Pigs were treated with transcutaneous BH targeting BME, The Hong Kong Polytechnic University, Hong Kong, China volumes of both kidneys using a 1.5 MHz FUS transducer operat- Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P84 ing at peak electric power ranging from 0.6 – 4kWunder coaxial B-mode ultrasound guidance. Sonication protocols delivered 10 OBJECTIVES Sonodynamic therapy (SDT), based on the synergis- pulses of 1-10 ms duration and 1% duty factor to 4 adjacent tic effect of low-intensity ultrasound and sonosensitizer, is a po- focal points spaced 1 mm apart without respiratory gating. Eu- tential noninvasive approach for the treatment of cancers. thanasia was performed immediately after treatment, and the However, the specific cellular death mechanisms of sonosensiti- treated regions of the kidney were carefully removed en bloc zers with different subcellular localization patterns remains un- and placed in ½ strength Karnovsky’s fixative. The tissue was known. In the present study, protoporphyrin IX (PpIX), either sliced into 1 mm sections parallel to the direction of treatment. endogenously induced by 5-aminolevulinic acid (ALA) or ex- Small regions from the identified lesions and their border ogenously administered, were used to investigate the effects of (1x2mm) were sampled and processed for transmission electron its subcellular localization pattern on key cellular activities in- microscopy (TEM). The remaining tissue slices were processed for cluding mitochondrial dynamics and cargo-selective pathways of histological evaluation. Semi-thin sections were taken from the autophagy in hela cells. TEM blocks and stained with 1% methylene blue to determine METHODS Different concentration and distribution patterns of from which locations to take the ultrathin sections. Ultrathin sec- ALA-induced PpIX and exogenous PpIX in Hela cells were deter- tions were double stained with uranyl acetate and lead citrate mined by microplate reader and confocal immunofluorescence and ultrastructural examinations were performed on JEM-100CX microscopy respectively. Cell viability, apoptosis, and autophagy and Zeiss Libra-120 transmission electron microscopes. were evaluated by microplate reader and flow cytometry. Mito- RESULTS On histologic assessment, lesions appeared to contain a chondrial dynamics and redox balance were examined by west- slurry of homogenized cellular debris without apparent cellular ern blot and confocal immunofluorescence microscopy. cargo- structures, areas of petechial hemorrhage, and a distinct lesion selective pathways of autophagy were determined by confocal border between treated and untreated kidney. At the ultrastruc- immunofluorescence microscopy. tural level, a border region between treated and untreated kidney RESULTS We demonstrated that both autophagy and apoptosis was observed measuring 10-20 microns, in which cell membranes existed in ALA-induced PpIX and exogenous PpIX mediated had been disrupted but intracellular organelles remained intact sonodynamic therapy. However, ALA-SDT mainly caused (Fig. 1). Within the lesion, ultrastructural analysis revealed regions mitochondria-specific autophagy by harming mitochondrial fu- of complete loss of structure with replacement of cells and or- sion and fission balance, exogenous PpIX-SDT caused non- ganelles by electron dense sub-micron cellular debris alternating selective autophagy by harming cytoplasmic proteins and or- with small areas of completely disrupted cells containing organ- ganelles. Furthermore, ALA-SDT caused unbalance of mitochon- elle ghosts. Throughout the lesions, fragmented collagen fibrils drial redox system by excessive production of mitoROS, while were observed (Fig. 1). Within lesions intact erythrocytes were exogenous PpIX-SDT caused unbalance of cytoplasmic redox observed within the slurry of cellular debris (Fig. 1) consistent system by excessive production of cytoplasmic ROS. Recipro- with post-treatment petechial hemorrhage. cally, MnTMPyP, mitoROS scavenger, rescued both cellular au- CONCLUSIONS This data represents the first ultrastructural study tophagy and apoptosis in Hela cells treated with ALA-mediated of BH lesions generated in an in vivo animal model. TEM sonodynamic therapy. evaluation revealed that aside from the presence of intact CONCLUSIONS Taken together, we conclude that different subcellu- erythrocytes, the lesion contents were similar in characteristic lar localization patterns of PpIX induce different cargo-selective path- to that observed in ex vivo tissue. At the border of the lesion ways of autophagy, serving as a pro-death function, cooperated with there is a gradation in ultrastructural changes not observed apoptosis to dictate the fate of cell death in sonodynamic therapy. microscopically. Funding: NIH R01 EB7643, K01 EB015745, NSBRI through NASA NCC 9-58, RFBR 16- 02-00653, and Ur- ology Care Foundation. P85 Focused ultrasound reprograms ethanol-treated prostate cancer cells back to normal Heng Yu, Hakm Murad, Daishen Luo, Damir Khismatullin Biomedical Engineering, Tulane University, New Orleans, Louisiana, USA Correspondence: Heng Yu Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P85 OBJECTIVES Prostate cancer is the most common and sixth deadli- est cancer in men worldwide. Since the majority of prostate cancer patients are elderly men, often not suitable for invasive procedures, there is a need for minimally invasive therapies (e.g., focused ultra- sound) against prostate cancer. Previous ex vivo and in vitro studies Fig. 1 (abstract P83). Representative TEM micrographs from the conducted in our laboratory showed that high-intensity focused center of the lesion (left), at the border of the lesion (center) and ultrasound (HIFU) synergistically enhanced tumor destruction by lower magnification view of the lesion border (right). The lesion ethanol injection. The objective of this study was to investigate the contained intact erythrocytes (E) in a slurry of cellular debris, molecular mechanisms behind anti-cancer effects of HIFU. Specific- fragmented collagen fibrils (arrowhead), and some ghosts of organelles ally, we tested the hypothesis that focused ultrasound drastically re- (G). At the border, intact but damaged organelles (O) lie in between duced the metastatic potential of chemically-treated cancer cells via cellular debris (D) and intact cells (C). At lower magnification (right), the overproduction of heat-shock protein 70 (HSP70), death receptor border region between fully intact and fully fragmented tissue is clearly Fas, its ligand FasL and TNF-α receptor. observed (yellow dotted lines) and measures approximately 20 METHODS DU-145 and PC-3 human prostate cancer cells were microns (black arrow) cultured in T-175 flask in full growth medium. The suspension of Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 104 of 122 cultured cancer cells (100 μl, 2.7million cells/ml) was placed in a P86 0.2 ml thin-wall PCR tube. The cells were left untreated (control) Therapeutic effect of focused ultrasound combined with dendritic or exposed to HIFU alone, 4% ethanol, or HIFU + 4% ethanol. cell treatment for melanoma: preliminary study The focused ultrasound signal was generated by a 1.1 MHz trans- Eun-Joo Park, Yun Deok Ahn, Yuri Cheon, Jae Young Lee ducer in the continuous or pulsed mode, with acoustic power Radiology, Seoul National University Hospital, Seoul, Korea ranged from 4.1 W to 20.52 W. HIFU level 4(H4) in this experi- Correspondence: Eun-Joo Park ment has the acoustic power of 8.72 W. The Fas, FasL and TNF-α Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P86 expression in the cancer cells was measured by flow cytometry at 2, 24, and 72 hours post-treatment. The Hsp70 protein levels OBJECTIVES As focused ultrasound (FUS) combined with microbubbles were determined by Western blot analysis at 2, 12, 24, 48 and 72 has been widely studied in cancer treatment, there is growing interests hours post-treatment. To confirm that cancer cells treated with in immunotherapy combined with FUS as FUS treatment might trigger ethanol and then HIFU lose their aggressiveness, we conducted a the immune response so that it results in therapeutic effects for cancer. series of adhesion and spheroid culture experiments with treated To investigate the effects of FUS on the body’s immune system, this cancer cells. Specifically, we measured the number of cancer cells study was designed as a preliminary study that evaluates the thera- adhered to TNF-α-activated endothelium under static conditions peutic effects of antigen-pulsed dendritic cells (DC) with FUS treatment. and tested the ability of the cells to form multi-cellular spheroids METHODS A total of 30 mice were inoculated with B16-F10 melanoma cells using a hanging drop method. The data (mean ± SEM) are the CFPAC-1 and divided into five treatment groups: control, antigen-pulsed DC result of 3-4 independent experiments. Statistical significance was only (DC), FUS only (FUS), and DC with FUS at two different FUS operating determined by Student’st-test. conditions DC-FUS1, DC-FUS2). Animals were treated on a weekly basis for RESULTS Prostate cancer cells, treated or not with ethanol, signifi- three weeks and posttreatment monitoring was followed for one week. cantly increased their expression of FasL immediately after being ex- RESULTS Animals in the group treated with DC and FUS2 which has posed to HIFU and continued to produce this molecule at a higher mechanical index (MI) and short duty cycle (5%) showed slower significantly higher amount than untreated (control) or ethanol tumor growth incomparison to control and DC only groups. Tumor alone-treated cells at 24 hours and 72 hours post-treatment (Fig. 1A). growth in the group treated with FUS only also showed lower growth The highest expression of FasL was achieved for the combined treat- rate than DC only and control groups. However, the group treated with ment group. Similarly, the combined treatment led to significantly DC and FUS1 which has lower MI and long duty cycle (50%) showed fas- higher expression of Fas than any other treatment at both 24 and 72 ter tumor growth than DC only, DC and FUS2, and FUS2 only groups. hours (Fig. 1B). The expression of TNF-α receptor was also signifi- CONCLUSIONS Based on the result that FUS treatment with high MI cantly increased in treatment group at both 24 and 72 hours (Fig. and very short duration can enhance therapeutic effects of DC treat- 1C). HSP70 expressed significantly higher in cancer cells exposed to ment, it is assumed that mechanical effects of FUS might be the main HIFU than that in untreated or ethanol alone-treated cells (Fig. 1D). mechanism for enhanced treatment effects of DC therapy. In order to Our static adhesion assay showed that the cancer cells in the com- investigate the detail of FUS effects on different therapeutic outcome bined treatment group attached much rarely to TNF-α-activated vas- and to improve the treatment protocol for enhanced therapeutic ef- cular endothelium than the cells in other groups. The cells exposed fects of the combined treatment further study will be followed. to both ethanol and HIFU were unable to form three-dimensional tumor spheroids. CONCLUSIONS Although Hsp70 plays a key role in cancer initiation and progression, its overproduction interferes with NF-κB signaling, P87 thereby causing apoptosis and reduced expression of adhesion mole- The long-term fate of the sonoporated pancreatic cancer cells is cules required for metastasis. Here, we showed that focused ultra- uncorrelated with the degree of the sonoporation sound induces Hsp70 overproduction in prostate cancer cells and 2,1 1 3 2 1 Lizhou Lin , Caixia Jia , Alfred C. Yu , Lianfang Du , Peng Qin promotes cell apoptosis via an increase in expression of Fas, FasL Department of Instrument Science and Engineering, Shanghai Jiao and TNF-α receptor. All these factors lead to phenotypic changes in Tong University, Shanghai, China; Department of Ultrasound, Shanghai surviving cancer cells that reduce their aggressiveness. First People’s Hospital, Shanghai Jiao Tong University, Shanghai, China; Department of Electrical and Computer Engineering, The University of waterloo, Waterloo, Alberta, Canada Correspondence: Lizhou Lin Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P87 OBJECTIVES Sonoporation, refers to the reversible membrane perfor- ation induced by the acoustic cavitation, has showed great potential for macromolecule delivery. Previous studies have determined the fate trend of the sonoporated cell. However, it is commonly believed that the sono- poration appeared to be heterogeneous due to discrete cavitation events adjacent to every single cell. The relationship between the long-term fate trend and the degree of the sonoporated cells is not still clear. This work aims to revealthe long-term fate of heterogeneously sonoporated cells. METHODS The pancreatic cancer Panc-1 cells mixed with 1% SonoVue microbubbles and 40 KDa FITC-dextran were placed in a tissue mimick- ing chamber and exposed to 1-MHz ultrasound with 1 kHz pulse repeti- tion frequency, 300 cycles and 0.6 MPa, and peak negative pressure (Fig. 1). Another 7.5MHz focused transducer was employed to passively detect the inertial cavitation dose (Fig. 2). The samples were firstly stained by Propidium Iodide at 30 min after exposure. After the sono- porated cells were identified by flow cytometry analysis, according to the relative fluorescence intensity (weak, medium and high) of the in- Fig. 1 (abstract P85). A: FasL expression in prostate cancer cells at ternalized FITC-dextran, the sonoporated cells were categorizedinto 2, 24 and 72 hours post treatment. B: Fas (CD95) expression in three sub-groups by flow cytometry sorting. After cultured for 6 H and prostate cancer cells at 2, 24 and 72 hours. C: TNFα expression in 24 H and stained by Annex V-alex350 and Propidium Iodide, these prostate cancer cells at 2, 24 and 72 hours post treatment. D: HSP70 three sub-groups ofsonoporated cells were detected to determine the expression in prostate cancer cell at 2 and 24 hours apoptotic and necrotic ratio by flow cytometry analysis. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 105 of 122 RESULTS The sonoporated cells were categorized into three sub-groups (sub-groups 1-3), which represented different degrees of sonoporation re- spectively (Fig. 3). The fluorescence intensity of the internalized FITC-dex- tran in sub-groups 2 and 3 was approximate 5.62-fold and 19.53-fold higher than that in sub-group 1, respectively. The apoptotic and necrotic ratio in all three sub-groups of the sonoporated cells gradually increased with the increasing culture time, compared with those in the control cells (Figs. 4 and 5). No significant difference in both the apoptotic (P > 0.05) and necrotic (P > 0.05) ratio were observed in three sub-groups of sonoporated cells which were cultured for 6 h and 24 h culture, respect- ively (at 6 H post-exposure, 2.48±1.65%,5.21±1.01 %, 5.05± 1.17 % and 5.95± 1.04 % apoptosis in the control and sub-groups 1-3 sonoporated cells, respectively. At 24 hour post-exposure, 4.40+2.81%,11.15± 2.48 %, 11.17± 4.93 % and 13.21± 3.78 % apoptosis in the control and sub- groups 1-3 sonoporated cells, respectively). CONCLUSIONS Our results indicated some of the sonoporated cells would experience apoptosis and necrosis in the long-term after inertial cavitation. The apoptotic and necrotic ratio in the sonoporated cells exhibited no sig- nificant correlation with the degree of sonoporation. These findings prelimin- arily suggest that the signal, that triggers the apoptosis and necrosis of the sonoporated cells, may be not correlated with the physical damage caused by acoustic cavitation, but dependent on underlying cellular response. Fig. 4 (abstract P87). Three sub-groups (1-3) of the sonoporated cells were respectively cultured for 24 H after inertial cavitation, then stained by Annexin V-Alex 350 and PI, and detected by flow cytometry analysis (A) Viable cells, (B) Sub-group 1 sonoporated cells; (C) Sub-group 2 sonoporated cells; (D) Sub-group 3 sonoporated cells Fig. 1 (abstract P87). Experimetnal Setupfor ultrasound exposure and cavitation detection Fig. 2 (abstract P87). Panc-1 cells mixed with 1% SonoVue bubbles (black water) and water (as control, red curve) exposed to 0.6 MPa PNP, 1 KHz PRP and 30% duty cycle. Frequency domain of the recorded signal in a period (2-1). Power of broadband components Fig. 5 (abstract P87). The apoptosis ratio of three sub-groups as a funtion of exposure time (2-2) sonoporated cells at 6 H and 24 H after ultrasound exposure P88 Tissue erosion at the fluid interface by dual-frequency HIFU excitation Yisheng Lei, Yufeng Zhou Nanyang Technological University, Singapore, Singapore Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P88 Fig. 3 (abstract P87). The sonoporated cells were catergorized into OBJECTIVES High-intensity focused ultrasound (HIFU) burst has been three sub-groups (1-3) according to the fluorescence of the used to successfully erode the tissue or gel phantom at the interface internalized FITC-dextran. A. the fluorescence of the control; B. The of fluid. The performance of histotripsy or microtripsy [high peak fluorescence of the reversible sonoporated cells. C. the relative FITC negative pressure, p-, and pulse repetition frequency (PRF > 100 Hz) of the three sub-groups 1-3 but short pulse duration (in the order of ms)] and those conventional Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 106 of 122 HIFU bursts [moderate p-, long pulse duration (in the order of ms), CONCLUSIONS The results of this study suggest that histotripsy has and low PRF (a few Hz)] have been confirmed. In order to further potential as a completely non-invasive liver cancer ablation method. understand the mechanism of tissue erosion and improve the cap- Future work is ongoing to investigate long-term safety and biological ability, a new technology, using dual-frequency excitations, was pro- response to histotripsy in relevant in vivo liver cancer models. posed and investigated in this study. METHODS The effect of frequency difference and modulation depth in the dual-frequency excitation on the produced erosion area and volume were quantitatively evaluated and compared to those of sin- gle-frequency excitation. In addition, the outcomes at the different PRFs were also compared. Bubble dynamics using different excitation strategies were captured by high-speed photography and passive cavitation detection (PCD). The acoustic field of dual-frequency exci- tation was simulated numerically and measured experimentally. The corresponding bubble oscillation was also simulated with the Gil- more model. RESULTS Dual-frequency excitation is more effective to disintegrate the gel phantom and tissue than single-frequency excitation up to about 2 fold. CONCLUSIONS In summary, this strategy can enhance the tissue ero- sion using the same output power, and operational parameters should be optimized for the bestoutcome. In vivo experiment will be carried out for the clinical translation. P89 Fig. 1 (abstract P89). (A) A custombuilt small animal system with a Non-invasive liver cancer ablation using histotripsy in an in vivo (B) 1 MHz therapy transducer was used to apply histotripsy to Hep3B murine model tumors in an in vivo murine model. During treatment, the (C) bubble Eli Vlaisavljevich, Tyler Gerhardson, Joan Greve, Shan Wan, Kim Ives, cloud and (D) tissue fractionation was visualized in realtime. After Timothy Hall, Theodore Welling, Zhen Xu treatment, histological analysis showed precise lesions generatedinside University of Michigan, Ann Arbor, Michigan, United States the tumor Correspondence: Eli Vlaisavljevich Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P89 OBJECTIVES Current liver cancer ablation methods are primarily ther- mal-based and thus share inherent limitations such as inconsistent ablation in tissue with nonuniform heat dissipation patterns or in- P90 complete tumor necrosis near major vessels. Recently, our group has Comparison of ultrasound-guided high intensity focused developed histotripsy as a completely non-invasive liver cancer abla- ultrasound and surgery for abdominal wall endometriosis: a tion method. Histotripsy is a non-thermal ultrasound ablation retrospective cohort study 2 1 2 method that fractionates tissue through the precise control of acous- Ling Zhao , Jinyun Chen , Chunquan Zhao tic cavitation. Previous experiments in an in vivo porcine model have College of Biomedical Engineering, Chongqing Medical University, shown the feasibility of using histotripsy to noninvasively create well- Chongqing, China; Department of Obstetrics and Gynecology, The 1st confined lesions in the liver through the intact chest, with sharp Affiliated Hospital of Chongqing Medical University, Chongqing, China boundaries between treated and untreated tissue. In this study, the Correspondence: Ling Zhao feasibility of using histotripsy for non-invasive liver cancer ablation Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P90 was tested in an in vivo murine model. METHODS Liver cancer patient-derived xenografts (Hep3B HCC cell OBJECTIVES To compare the safety and efficacy of high-intensity fo- line) were subcutaneously implanted in 8 NSG mice. When tumors cused ultrasound (HIFU) and surgery for abdominal wall endometriosis. reached ~1 cm, histotripsy was applied non-invasively using a custom- METHODS With a retrospective cohort study design, fifty-four Chinese built 1 MHz histotripsy therapy system designed for small animal exper- women with abdominal wall endometriosis who underwent ultrasound- iments (Fig. 1A/B). Guided by real-time ultrasound imaging, histotripsy guided HIFU between January 2012 and December 2014 were enrolled. was applied to the tumors using 1-2 cycle pulses, a pulse repetition fre- In which, 29 cases included in surgery group, 25 cases in HIFU group. quency of 100 Hz, and an estimated in situ peak negative pressure >30 Technological success, adverse events and recurrent rate were assessed MPa.The targeted tumor volume was mechanically scanned using a ro- and compared between two groups. botic micro-positioner controlled by a PC console. After treatment, le- RESULTS The clinical features are comparability, and the success rate sion characteristics were assessed using ultrasound imaging and MRI was obtained of 100% both in the two groups. The clinical efficacy (7T small animal scanner), and the treated tissues were then harvested rate was 92% (22/25) in theHIFU group, and 100 % (29/29) in the sur- for gross analysis and histological evaluation. All procedures were ap- gery group. The numeric rating scales (NRS) after HIFU were signifi- proved by the ICUCA at the University of Michigan. cantly lower than those before the procedure from 6.92 to0.28 RESULTS Histotripsy-induced cavitation clouds were successfully (P<0.05). During the mean follow-up of 32 months (range: 19-46 generated inside the tumors of all subjects, with the bubble clouds months), the duration pain relief were 39.00±30.02 months in the clearly visualized as a hyperechoic zone on ultrasound imaging surgery group and 30.96±28.55 months in the HIFU group (P>0.05). (Fig. 1C). Immediately after treatment, the histotripsy-induced tissue Three cases (10.71%) experienced recurrence of pain in the surgery damage was visualized as a hypoechoic zone on ultrasound imaging group, and 2 (8.00%) in the HIFU group. Adverse events occurred in (Fig. 1D). Gross and histological analysis demonstrated that histotripsy 4 (13.79%) surgery cases and in 1 (4.00%) of HIFU cases respectively, resulted in the complete fractionation of all tumor cells within the tar- without significant difference (P>0.05). Events in the surgery group geted region into an a cellular homogenate with no remaining cellular included incision healing delayed and lung infection, and skin burn structures and sharp boundaries (<10 μm) between treated and un- occurred in the HIFU group. treated tissue (Fig. 1E). No gross damage to the surrounding tissue was CONCLUSIONS HIFU appears to be safe and effective for the treat- observed in any subjects. ment of abdominal wall endometriosis. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 107 of 122 P91 Prediction of thermal lesion formation by short-pulse pre-exposure for cavitation-enhanced ultrasonic heating 1 2 2 1 Ryosuke Iwasaki , Ryo Takagi , Shin Yoshizawa , Shin-ichiro Umemura Biomedical Engineering, Tohoku University, Sendai, Miyagi, Japan; Communications Engineering, Tohoku University, Sendai, Miyagi, Japan Correspondence: Ryosuke Iwasaki Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P91 OBJECTIVES Acoustic cavitation microbubbles are known to enhance the heating effect of ultrasound. In high-intensity focused ultrasound (HIFU) treatment, utilizing microbubbles is expected to accelerate ultrasonic heating to reduce treatment time. However, it is not sim- ple to control the position of generating cavitation bubbles. There- fore, it is necessary to visualize either the bubbles or the effective focal zone where HIFU is most absorbed. If such visualization is per- formed in prior to the cavitation enhanced HIFU treatment, both safety and effectiveness of the treatment will be ensured. The object- ive of this study is to compare the methods between visualizing the ultrasonic backscatter based on B-mode images and the visualizing the ultrasonic attenuation based on acoustic radiation force impulse (ARFI) technique. Fig. 1 (abstract P91). Schematic of experimental set up and HIFU METHODS Figure 1 shows the experimental setup and the sequences sequence as push pulse and therapeutic beam of HIFU exposure for pre-treatment focal zone visualization and tis- sue coagulation. A piezocomposite 2D-array HIFU transducer with both aperture diameter and geometrical focal length of 120 mm was placed in a water tank and connected to a staircase drive amplifier. A sector diagnostic array probe was set in the central hole of the trans- ducer and connected to an ultrasound imaging system. For focal zone visualization, a high-intensity pulse called trigger pulse at an in- tensity of 60 kW/cm2 with a duration of 0.1 ms for generating cavita- tion bubbles was followed by a moderate-intensity HIFU burst at an intensity of 3 kw/cm2 with a duration of 1.9 ms for inducing dis- placements. A chicken breast was used as a sample tissue. Before and immediately after this push pulse exposure, RF data were ac- quired via high-speed ultrasound imaging. The distribution of axial displacements between before and after push pulse exposure was calculated from the phase shift in 1D cross-correlation. After that, tis- sue was coagulated by the repetition of the trigger pulse followed by a HIFU burst with a duration of 44.9 ms, which was continued for 6 s with a duty cycle of 90%. The resulting tissue coagulation was compared with the distribution of the HIFU induced displacement and the B-mode image. RESULTS Figure 2 shows the distributions of HIFU induced displace- ments without and with the trigger pulse, the subtraction B-mode images, and the gross pathologiesof the coagulated tissue. The re- gion of the heat source seems to have been shifted backward from Fig. 2 (abstract P91). Distributions of axial displacements (b), (f), the HIFU focal point. With the trigger pulse, HIFU seems to have subtraction Bmode images (c), (g) and slices of samples after been shielded by the cavitation bubbles. To test this, an offset of coagulation (d), (h) without and with a focal depth offset of 7mm, 7 mm was given to the focal length of the trigger pulse. By respectively, and without trigger pulses following push pulses (a), (e) comparing Fig. 2(b)-(d) and Fig. 2(f)-(h), it is obvious that the off- set extended the area of coagulation deeper, which was well pre- dicted by the effective focal zone by the HIFU induced displacements. The regions of high brightness in the subtraction B-mode images were thought to be the regions where cavitation P92 bubbles were generated. Research on EFT immunity of ashi ultrasonic therapeutic apparatus based on statistical analysis CONCLUSIONS In this study, we proposed HIFU induced ARFI im- Zhiming Zhong, Fangwei Ye, Yang Liu aging employing ultrasonic high-speed imaging to visualize the ef- Biomedical Engineering Collage, Chongqing Medical University, fective focal zone of HIFU. The results showed that the proposed Chongqing, China method was effective for prediction of thermal lesion formation in- Correspondence: Zhiming Zhong duced by HIFU heating even under difficult situation with enhance- Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P92 ment by cavitation bubbles. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 108 of 122 OBJECTIVES Ashi ultrasonic therapeutic instrument can effectively al- CONCLUSIONS Certain intensity and exposure time of ultrasound leviate the chronic soft tissue injury pain, the EFT immunity influ- can enhance the diffusion effect of small beeds solution in the ences the therapeutic effect ofthe equipment. In order to diminish agarose gel. the randomness in the EFT immunity test, it is essential to adopt stat- istical analysis for research on EFT immunity of Ashi ultrasonic thera- peutic apparatus. METHODS The parameters of EFT generator were set according to test standard GB/T 17626.4 or IEC 61000-4-4, but the EFT initial voltage was set as 200V. The power line of Ashi ultrasonic therapeutic apparatus was injected with EFT. Then, the operative mode of equipment was ob- served as a evaluation criteria for judging whether the apparatus was disturbed. If not, the EFT voltage was increased with a step 200V, and continuing test after 30 seconds until that the equipment was dis- turbed. Then, one test was finished and the EFT voltage at this time was record, which was called EFT interference threshold voltage. Ac- cording to the above method, the EFT immunity test was carried out in many times for obtaining multiple EFT interference threshold voltage data which was used to solve unknown parameters in two parameter Weibull distribution model through Maximum Likelihood Method. The distribution law of EFT interference threshold voltage of Ashi was ana- lyzed by theprobability density function (PDF) and cumulative distribu- Fig. 1 (abstract P93). Control sample, no ultrasound applied. The tion function(CDF) curve of Weibull distribution model which has been Xdirection is along the ultrasound beam, the Yaxis is the fluorescent verified by chi-square goodness of fittest. intensity. The profile shows the crosssection the sandwich gel RESULTS According to the result of chi-square goodness of fit test, two parameter Weibull distribution model was available for research on dis- tribution law of EFT interference threshold voltage of Ashi ultrasonic therapeutic instrument. The PDF and CDF curves prove that EFT inter- ference threshold voltage of Ashi equal 900V, and the EFT with differ- ent voltage has different interference probability for Ashi. CONCLUSIONS The test standard GB/T 17626.4 or IEC 61000-4-4 proves that the standard voltage in the EFT immunity test equal 1000V which is the actual output voltage, the value greater than EFT interference threshold voltage of Ashi 900V. Therefore, the apparatus cannot pass EFT immunity test in great probability. The long data communication cable is coupled by EFT easily, which is main reason for Ashi cannot pass test probably. So, it is essential to shorten properly power line or data communication cable in the equipment. The statistical analysis is scientific and useful for obtaining EFT interference threshold voltage of equipment, which has important significance for the research on EFT immunity. Analyzing EFT immunity of equipment based on statistical analysis could decrease effectively randomness in the test result. And the distribution law of EFT interference threshold voltage of instrument will be analyzed accurately and clearly by statistical model. Fig. 2 (abstract P93). Control sample, no ultrasound applied. The Xdirection is along the ultrasound beam, the Yaxis is the fluorescent P93 intensity. The profile shows the crosssection the sandwich gel Experimental study of the promotion of diffusion process in a biofilm by low intensity ultrasound Dong Ma Physics, University of Vermont, Burlington, Vermont, United States Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P93 OBJECTIVES Experimentally to show the Low intensity ultrasound can dramatically enhance nanometer liposome penetration into the biofilm films. METHODS The biofilm model sample was made of three layers of 1% agarose gel; each layer has the thickness of 0.5 mm. The first layer of agarose contains fluorescent beads (diameter: 100 nm) was made in a petri dish, after the agarose was cooled down, a second layer was made on top of the first layer; at this point a two layer agarose gel was made. Flip it over, and made the third layer on top, then we had a three-layer gel with a fluorescent layer sandwiched between two non-fluorescent agarose gels. An ultrasonic tone burst (frequency=2.25 MHz, 10% duty cycle and spatially and temporally averaged intensity, ISATA 4.4 W/cm2) generated by a transducer of diameter 1.9 cm was used to treat the sample for 10 minutes. RESULTS When ultrasound was applied, the diffusion distance Fig. 3 (abstract P93). Ultrasound applied. The Xdirection is along (0.8 mm) is much longer than those without ultrasound (0.1mm). the ultrasound beam, the Yaxis is the fluorescent intensity. The profile Figures 1 and 2 show that there are steep drops on the right side shows the crosssection the sandwich gel (which we are interested in), Figs. 3 and 4 show a very different profile. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 109 of 122 because of the acute pain. The pain threshold was recorded when the muscle contraction at the first time. The pain threshold value in- creasing of ultrasound group were statistically significant compared with the control group on the 3rd and muscle contraction at the first time. The pain threshold value increasing of ultrasound group were statistically significant compared with the control group on the 3rd and 24th (P < 0.05), 7th, 10th and 17th day (P < 0.01) (Fig. 1).Inflam- mation levels: The levels of the 5-HT (Fig. 2A) and PGE-2 (Fig. 2B) in the injured hind leg muscle of rabbit were reduced largely in focused ultrasound irradiation group from the 3rd dayto 17th day after the first treatment (P < 0.05), and the IL-1β level appeared a decreasing trend until the 24th day from the first treatment (P < 0.05) (Fig. 2C). CONCLUSIONS It can be concluded that the focused ultrasound de- vice used in this study is safe and effective in the treatment of soft- tissue injury by cutting inflammation factor expression and increas- ing pain threshold. Fig. 4 (abstract P93). Ultrasound applied The Xdirection is along the ultrasound beam, the Yaxis is the fluorescent intensity. The profile shows the crosssection the sandwich gel P94 The effect of focused ultrasound to treat the soft tissue injury on rabbit models Dandan Liang Chongqing Medical University, Chongqing, China Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P94 OBJECTIVES To evaluate the efficacy of focused ultrasound in the treatment of soft-tissue injury and preliminary discussed mechanism. METHODS Ethics statement: The Ethics Committee of Animal Experi- ments at Chongqing Medical University (Production License No. SCXK 2012-0001) approved the experiment and animal care protocols. Animals: Male and female New Zealand White rabbits (3-4 months old, 2.0–2.5 kg) were chosen from the Animal Center at Chongqing Fig. 1 (abstract P94). See text for description Medical University. All animals were housed in individual cages with a 12:12-h light–dark cycle and a temperature- controlled environ- ment (24 ± 2°C) and given a standard laboratory diet with drinking water at liberty. The animals were adapted to their environment at least 1 week before the experiment started. Soft-tissue injury model construction by hammer blow: As previously reported, the rabbits were fixed in the lateral and outside of the left hind was shaved by shaver and hit thrice on the thigh muscle by a cylindrical hammer (200 g in weight, 2.8 cm in bottom diameter), which fallen freely and vertically from inner of a hard smooth plastic tube (12 cm in length and 3 cm in inner diameter).The act repeated once every two days. When struck for four weeks and bred normally for three days later, the closed soft tissue injury was formed without femoral fractures. Intervention: All model rabbits were grouped into two groups ran- domly:ultrasound treatment group and non ultrasound group. The ultrasound group received a focused ultrasound treatment by a fixed and mobile method, which rabbits were treated fixedly for 20s, then taken a 5s break, lastly treated by mobile method with, 5–10 mm/s for 60s. They were treated per day for continuous ten days. Pain threshold determination: After treatment 30min, the pain threshold was measured by Analgesy-Meter. The rabbit was fixed and pres- Fig. 2 (abstract P94). See text for description sured gradually by the forcing bar and the value, the animal's pain threshold, was recorded when the animal retracted its leg. Measure- ment of PGE2,5-HT and IL-1β concentrations: The muscle 1g was ho- mogenized in 2ml phosphate buffered solution (PBS pH 7.4). Then P95 centrifuged and the supernatant was collected to detect the prosta- Detection of harmonic signal of the focused ultrasound based on glandin-E2(PGE-2),5-hydroxytryptamine(5-HT) and interleukin-1 beta acousto-optic effect (IL-1β) with an enzyme immunoassay at the 1st, 3rd, 7th, 10th, 17th, Fu Li, Hua Wang 24th days after the first of treatment. The parameter was measured Chongqing Medical University, Chongqing, China according to the protocols of respective kits (Shanghai Hushang Bio- Correspondence: Fu Li engineering Institute, China). Statistical analysis: Results were Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P95 expressed as mean ± standard deviation (SD). SPSS 19.0 (SPSS Inc., Chicago, IL, USA) was used for all statistical analyses. Independent sam- OBJECTIVES To explore a non-invasive method for detecting the har- ples t-test was used to analyze all experimental data. P < 0.05 (95% monic signal of high intensity focused ultrasound (HIFU) confidence interval) was considered statistically significant. METHODS When a parallel laser beam passed through the focus of the RESULTS Pain threshold value: When the stimulation intensity was in- focused ultrasonic field, the light signal was converted into an electrical creased gradually, the rabbits would show muscle contraction signal by a photodiode in the distance, and the electrical signal was Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 110 of 122 simulated by the Matlab. To verify the theoretical model, a focused P97 ultrasonic transducer was used as the source of ultrasonic emission, The value of the parameters in the peripheral blood routine test and it was placed in degassed water. A He-Ne laser was used to trans- for the preoperative diagnosis of uterine sarcoma: a review of a mit beam crossing the focus of the sound field, and the optical signal multicentre study in western China whose crossed focus was received by a photodiode (it was connected Dan Li, Wenzhi Chen, Jinyun Chen to the photoelectric detection circuit) in the distance, the output signal College of Biomedical Engineering, Chongqing Medical University, of the photoelectric detection circuit was collected by a digital oscillo- ChongQing, China scope, and the spectrum of the signal was obtained by Matlab (Fig. 1). Correspondence: Dan Li Comparison the normalized frequency spectrum of theoretical and ex- Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P97 perimental obtained. RESULTS The normalized frequency spectrum obtained from the simu- OBJECTIVES The current study was to examine the accuracy of pre- lation of the theoretical model is consistent with the experiment results. operative diagnosis of uterine sarcomas in western China and to evalu- CONCLUSIONS The method proposed in this paper is feasible to de- ate the efficacy of the indicators in the peripheral blood routine test for tect the harmonic signal of HIFU field, and it also lays a theoretical the differential diagnosis between uterine sarcomas and leiomyomas. and experimental foundation tomeasurement the sound pressure of METHODS A total of 102 patients with uterine sarcoma were evaluated, the nonlinear HIFU at focus. underwent surgery in the first affiliated hospital of Chongqing Medical University, the DapingHospital affiliated to the Third Military Medical University, People’s liberation Army of China and the affiliated Hospital of Southwest Medical University, covering from January 1st 2010 to De- cember 1 st 2015. Meanwhile,119 women with leiomyoma, complete clinical and pathological information documented at the time of sur- gery were selected as controls between January 1st 2010 to December 1 st 2015. Study parameters included age at the time of surgery, clinical findings, blood test results, imaging examinations results (specifically ultrasonography and magnetic resonance imaging [MRI]), endometrial cytology findings, postoperative pathological diagnosis and follow-up. Receiver operating characteristics (ROC) analysis was used for specificity and sensitivity estimates. The resulting area under the curve (AUC) indi- cates the average sensitivity of a marker over the entire ROC curve. Fig. 1 (abstract P95). The normalized frequency spectrum obtained Multivariate analysis was performed using nonlinear model of Logistic from the simulation of the theoritical model is consistent with the regression analysis and partial leastsquares-discriminant analysis. experiment results RESULTS At the final preoperative diagnosis, 59.8% (61/102) of uter- ine sarcomas were diagnosed as having malignant disease. 6 indexes including the WBC, neutrophil, monocyte, NLR, MLR and NWR had a certain diagnostic value evaluated by the ROC curve (AUC>0.70) (Fig. P96 1). A comprehensive system combined with by six markers for identi- Effect of cavitation induced by High Intense Focused Ultrasound fication of uterine sarcoma were analyzed by the ROC curve,non- (HIFU) on tungsten filament and stainless steel filament at 10 MPa linear model of Logistic regression analysis and partial least squares- static pressure discriminant analysis (Fig. 2).The AUC of this comprehensive diagno- 1,2 1,2 2 2 2 Yurong Zhang , Faqi Li , Xiaobo Gong , Qi Wang , Guangrong Lei , sis system was 0.90 (95%CI, 0.86to 0.94; sensitivity =76.1%,specificity 1,2 Zhibiao Wang =89.1%).Logistic regression analysis and partial least squares-discrim- State Key Laboratory of Ultrasound Engineering in Medicine Co- inant analysis showed that the comprehensive system combined founded by Chongqing and the Ministry of Science and Technology, with by six markers had a better value of differential diagnosis be- College of BiomedicalEngineering, Chongqing Medical University, tween uterine sarcomas and leiomyomas. Chongqing, China; National Engineering Research Center of Ultrasound CONCLUSIONS The comprehensive system combined with by six Medicine, Chongqing, China markers (WBC, neutrophil, monocyte, NLR, MLR and NWR) in the per- Correspondence: Yurong Zhang ipheral blood routinetest, which is convenient, reproducible, and in- Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P96 expensive, had a potential value of differential diagnosis between uterine sarcomas and leiomyomas (Fig. 3). OBJECTIVES To illustrate the intense cavitation and high energy density induced by spherical cavity transducer with open ends. METHODS We processing on tungsten filament and stainless steel fila- ment with the same diameter based on the extreme transient condi- tion, such as high pressure and high temperature during cavitation induced by continuous high intense focused ultrasound (CHIFU) and pulsed high intense focused ultrasound (PHIFU) at 10 MPa static pres- sures. The images are taken using a high speed camera, and scanning electron microscope (SEM) fractographic analysis is also conducted. RESULTS The 0.2 mm diameter tungsten filament and stainless steel filament were severed by CHIFU (6000 W of electric power) cavitation within 1.39 s and 0.56 s, respectively. PHIFU (1,5000 W of electric power) cavitation could severed 0.2 mm diameter tungsten filament within 0.22 s. SEM fractographic analysis showed that the fracture of stainless steel filament was general fatigue fracture, but the fracture of tungsten filament was relatively complex, which might be in- volved annealing andrecrystallization texture. CONCLUSIONS The tungsten filament is more difficult to be severed by acoustic cavitation than stainless steel filament, and HIFU can effect- ively improve plastic of tungsten filament. Next, we hope to calculate Fig. 1 (abstract P97). Comparison of differential peripheral blood the pressure and temperature of cavitation core region, known as hot- parameters by the receiver operating characteristic curve. AUC>0.70 spot, from the perspective of micro fracture mechanicsto illustrate the was considered to have a certain diagnostic value high energy density of our system. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 111 of 122 METHODS Figure 1 shows the measurement setup of the optical phase contrast method. Ultrasound pressure field in water creates a modulation of the refractive index, which further modulates the phase of the light passing through the field. The non-diffracted com- ponent of the light was focused exactly at the focal point of the sec- ond convex lens, while the other component diffracted due to the phase modulation was laterally slightly away from the focal point. By shifting the phase of the non-diffracted component by the phase plate typically by 90 degrees, the phase modulation of the diffracted component was converted to amplitude modulation through inter- ference with the phase-shifted non-diffracted component and then measured by the camera. The numerical calculation steps based on the measurement were as follows,1, An upstream field in which the optical phase does not wrap and the effect of nonlinear propagation can be ignored was chosen.2, The 3D pressure field was recon- structed from the projected 2D data from measurement by a CT al- gorithm.3, Nonlinear ultrasonic propagation was simulated using the obtained upstream pressure field as the input. In this study, an axi- symmetric 8-element annular array transducer (outer diameter: 80 mm, inner diameter: 36 mm, focal length: 80 mm, center frequency: 1.4 MHz) wasdriven at 65.6Vpp, the pressure field 40 mm upstream the focal point was measured, and the numerical simulation based Fig. 2 (abstract P97). Receiver operating characteristic curve on a pseudo spectral method was performed. The pressure ±5 mm analysis for combined with six markers from the focal point was also measured by scanning a fiber optic hydrophone (Onda, HFO-690) to compare with the proposed method. High-intensity focal field at a therapeutic level can be calcu- lated by multiplying the measured low-intensity pressure before in- putting it to the simulation of nonlinear propagation. Here, it was assumed that the nonlinear propagation at the upstream of the mea- sured area can be ignored even when the acoustic pressure is multiplied. RESULTS Figure 2 (a) shows the optically measured axisymmetric pulsed pressure field upstream the focal point, and Fig. 2 (b)-(e) show the results from the numerical simulation using it as the input. Figure 3 shows the comparison of the obtained focal field between hydro- phone scanning and the proposed method. The pressure distribution in axial and lateral directions is shown. The absolute positive and negative pressures measured by hydrophone were 2.89 and -1.75MPa, whereas they were 2.72 and -2.03MPa determined by the Fig. 3 (abstract P97). Partial least squares-discriminant analysis optical measurement. Good agreement was observed for the wave- (PLS-DA) combined with six markers forms and absolute pressure in both directions. CONCLUSIONS Pulsed ultrasound pressure field was quantified by numerical acoustic holography based on a fast optical measurement using a phase contrast method in combination with a CT algorithm. P98 Both pressure waveform and absolute pressure from the proposed Quantitative measurement of high-intensity pulsed ultrasound method agreed well with that from hydrophone scanning. Further pressure field using combination of optical phase contrast method studies underway are applications to high pressure field at a and acoustic holography therapeutic level and to continuous wave field such as that of HIFU 1 2 2 2 Takuya Nakamura , Hiroki Hanayama , Ryo Takagi , Shin Yoshizawa , (high-intensity focused ultrasound). Shin-ichiro Umemura Biomedical Engineering, Tohoku University, Miyagi, Japan; Communication Engineering, Tohoku University, Miyagi, Japan Correspondence: Takuya Nakamura Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P98 OBJECTIVES In recent years, ultrasound has been widely used for therapeutic as well as diagnostic purposes. To assure safety and effi- cacy of such application, accurate measurement of ultrasonic pres- sure field is necessary. Hydrophone scanning is the most common method to measure ultrasound pressure field. However, this method requires very long scanning time and has the risk of disturbing pres- sure field. In this study, pulsed ultrasound pressure field was quanti- fied by numerical acoustic holography based on a fast optical measurement using a phase contrast method in combination with a Fig. 1 (abstract P98). Optical phase contrast measurement setup CT algorithm. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 112 of 122 intensity and surface temperature is determined by the boundary condition of those two interfaces. Boundary condition of two inter- faces is investigated theoretically in this study. METHODS At the absorber/water interface, acoustic streaming generated by ultrasound lash the interface. In the focal region, the ultrasound field distributed like a cylinder-shape, so in this paper, the flow heat transfer analogy for a single round jet stream. According the time-average energy conservation from the transducer to the focal region, the relationship between average jet velocity and heat transfer coefficient at absorber/water inter- face and focal intensity can be derived, as show in Figs. 1 and 2. At the absorber/air interface, due to the ultrasound irradiation, the surface temperature elevated may cause the natural convection and radiant heat transfer. So the combined heat transfer can be expressed as ΦTotal =ΦConvection + ΦRadiation = AhcΔT+AhrΔT= AhTotalΔT, where htotal is combined surface heat transfer coeffi- cient, hc is convective heat transfer coefficient, hr is radiant heat transfer coefficient, T is the temperature elevation. The relation- Fig. 2 (abstract P98). Optically measured pulsed pressure field ship between transfer coefficient and temperature change is upstream the focal point (a), and the numerically simulated fields (b) shown in Fig. 3. 8.6 µs (c) 17.1 µs (d) 25.7 µs after the input, and(e) at the focus RESULTS According to the boundary analysis, the average heat trans- fer coefficient at absorber/water interface has no significant impact to the absorber/air temperaturechanges, for absorber's thicknesses from 1~4 mm, as shown in Table 1. When the surface temperature changes below 100 °C, the results from combined surface heattrans- fer and heat transfer was closed, the difference of maximum temperature elevation and temperature change rate less than 0.7%, as show in Table 2. CONCLUSIONS According the study, the boundary condition at water/absorber interface can be treated as infinite and constant temperature boundary conditions, when the heating time is very short. And at absorber/air interface, the heat conduction plays a major role, when the interface temperature changed below 100 °C. Fig. 3 (abstract P98). Comparison of pressure waveform between proposed method and hydrophone scanning in axial (a) and lateral (b) directions P99 Boundary analysis of absorber irradiated by the focused ultrasound 1 2 1 1 Ying Yu , Guofeng Shen , Chunlei Lei , Helin Zhang School of Computer, Jiangxi University of Traditional Chinese Medicine, Nanchang, Jiangxi, China; School of Biomedical Engineering, Shanghai Jiao TongUniversity, Shanghai, Shanghai, China Correspondence: Ying Yu Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P99 OBJECTIVES Characterization of focused ultrasound field is important factor for both efficacy and safety of clinical treatment. Recently, the infrared imaging has been used to estimating the intensity of fo- cused acoustic field. The principle of this method is estimating the absolute intensity and relative distribution of focused ultrasound by the temperature elevation at the surface of an absorber which irradi- ated by the focused ultrasound. There were two interfaces, absorber/ air and water/absorber, with large differences in thermal physics Fig. 1 (abstract P99). See text for description properties. In theory, the derived relationship between incident Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 113 of 122 P100 An inverse method for estimating the acoustic intensity in the HIFU field by infrared thermometry 2 1 1 1 Guofeng Shen , Ying Yu , Chunlei Lei , Helin Zhang School of computer, Jiangxi University of Traditional Chinese Medicine, Nanchang, Jiangxi, China; School of Biomedical Engineering, Shanghai Jiao Tong University, Shanghai, Shanghai, China Correspondence: Guofeng Shen Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P100 OBJECTIVES The 3D acoustic field distribution of HIFU is not only the key parameters for evaluating the quality of the transducer, but also an important indicator of the security and efficiency of HIFU system. Recently, a new method which based on infrared (IR) imaging was in- troduced. Authors (A. Shaw, et al and M. R. Myers, et al) have estab- lished the relationship between absorber surface temperature and incident intensity during the absorber was irradiated by the trans- ducer. Theoretically, the shorter irradiating time makes estimation more in line with the actual results. In this study, an inverse method was introduced to estimating the acoustic intensity of HIFUfield using the surface temperature (Fig. 1). METHODS The physical definition of this inverse problem is to recon- struct the unknown incident intensity distribution by measuring the thermal field of the absorber asa function of time. So, the thermal field of the absorber was measured with the IR camera not only dur- ing the heating time but also including the cooling time and pre- heating time. The algorithm for the solution consists of 9 steps.1) Fig. 2 (abstract P99). See text for description choose an initial guess2) solve the direct problem to obtain the sur- face temperature 3) solve the adjoint problem to find the gradient to the function J' 4) compute the conjugate coefficient 5) compute the search direction 6) solve the sensitivity problem with input data 7) compute the step size in the search direction 8) compute the new estimate 9) interrupt the iterative procedure if the stopping criterion is satisfied. Otherwise repeat the iteration until convergence is achieved. RESULTS The method proposed for quantitative assessment of acoustic intensities using IR camera with inverse method has a satis- factory percentage difference, in both maximum intensity (< 13.73 %) and -6 dB beam width (< 10.04 %) in focal region, in comparison to the theoretical simulation using a three-layer medium model (Fig. 2). Fig. 3 (abstract P99). See text for description CONCLUSIONS Thepercentagedifferenceincreasewiththeheat- ing time with zero mean noise, but decrease with heating time when the noise can be ignored. Table 1 (abstract P99). The table of maximum temperature change rate with different average heart transfer coefficient at absorber/ water interface Table 2 (abstract P99). The table of maximum and the maximum temprature change rate with different average heat transfer coefficient at absorber/water interface and heart conduction Fig. 1 (abstract P100). Comparison of the inverse method (red and blue dot line) and theoretical simulation (blue solid line) for heating of 120 and 200 ms Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 114 of 122 Table 1 (abstract P100). See text for description RESULTS Sound field distribution of propagating and standing wave focusing (Fig. 1) There were obvious differences of sound field distri- bution of sound propagation axis between propagating and standing wave focusing. Comparation of nonlinear effects of the two focusing modalities on the same focal pressure (Fig. 2) The nonlinear ef- fect of propagating wave is stronger than that of standing wave because of the stronger distortion of waveform. Recording the formation of lesion in phanton with high-speed camera (Fig. 3) Lesion formation of the propagating wave focusing modalities has been formed the “ellipsoidal” lesion whose major axis was along with wave propagation direction (Fig. 3a). However, lesion formation of focusing of standing wave has been formed the “gourd string” lesion (Fig. 3b). The focal 3-6MHz broadband emis- sion during exposure collected with PCD on the same focusing pressure (Fig. 4) There have been no significant differences of the cavitation signal between the two focusing modalities. CONCLUSIONS 1. Lesion formed in the condition of standing wave focusing is obviously different from that in the condition of propagating wave2. Focusing modality of standing wave need lower power than that of propagating wave to get the same acoustic pressure3. Nonlinear effect plays an important role in the formation of lesion. Fig. 2 (abstract P100). Comparison of the inverse method (blue dot line) and theoretical simulation (red solid line) for heating of 120 ms Table 2 (abstract P100). The difference between the maximum intensity on-axis using the proposed method in the simulated theoretical prediction for different heating time with same noise Fig. 1 (abstract P101). Sound field distribution of three axes of propagating wave focusing (a) and standing wave focusing (b) P101 Comparation of the focusing ultrasound of propagating and standing wave in sound field and bioeffects 2,1 2,1 2,1 2,1 Man Luo , Faqi Li , Zonggui Chen , Yurong Zhang National Engineering Research Center of Ultrasound Medicine, Chongqing, China; State Key Laboratory of Ultrasound Engineering in Medicine Co-Founded by Chongqing and the Ministry of Science and Technology, College of Biomedical Engineering, Chongqing Medical University, Chongqing, China Correspondence: Man Luo Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P101 OBJECTIVES This study aims to investigate the differences of this focus- ing modality with the conventional one in sound field distribution and Fig. 2 (abstract P101). Acoustic pressure waveform of propagating the process of lesion formation under the same acoustic pressure. wave (a) and standing wave (b) on the same focusing pressure METHODS Setting two identical transducers in opposition whose internal (16MPa) surface were on the same spherical surface and driving simultaneously to achieve the focusing of standing wave. And then driving one transducer and sheltering another one from ultrasound beams with sound absorbing materials to get the focusing of propagating wave. The sound field distri- bution of the two focusing modalities were acquired by fiber-optic hydro- phone. The change of the acoustic pressure of focus with the increase of power were measured by fiber-optic hydrophone also and waveform of acoustic pressure were recorded to analyze nonlinear effects of the two fo- cusing modalities on the same focusing pressure (16MPa). Linear fitting of acoustic pressure squared with power extrapolate the powers of getting the experimental acoustic pressure (18MPa). Tissue mimicking phantom were exposed with the extrapolating powers of the two focusing modal- Fig. 3 (abstract P101). Recording the formation of lesion in ities. Meanwhile, the processes of lesion formation in phantom were re- phanton with highspeed camera; (a) the processes of lesion corded with high-speed camera to analyze the differences of lesion formation of the propagating wave focusing modalities; (b) the formation between the two focusing modalities and the focal 3-6MHz processes of lesion formation of focusing of standing wave broadband emissions were recorded with PCD to analyze cavitation. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 115 of 122 value is 0.47, while the values of the erosion treated by histo- tripsy distributed between 0.2 and 1.2 and the meanvalue is 0.40 (Fig. 2). From the results, we find that the mean values of the Nakagami parameter M for the thermal lesion and the mechanic- ally homogenized lesion were different. CONCLUSIONS The preliminary study on the tissue-mimicking phan- tom suggested that the Nakagami parameter M may have the poten- tial possible to identify the lesions treated by HIFU of different modes and realize the structure characterization for the different HIFU lesions in the tissues. Fig. 4 (abstract P101). Cavitation signal of the propagating wave focusing modalities (the black curve) and the standing wave focusing modalities (the green curve) on the same focusingpressure (18MPa) Fig. 1 (abstract P102). The images of the different HIFU lesions. P102 (a) Bmode image of the thermal lesion; (b) Nakagami image of the Feasibility of using nakagami distribution in structure thermal lesion; (c) Bmode image of the lesion byhistotripsy; (d) characterization of the different lesions treated by HIFU Nakagami image of the lesion by histotripsy Meng Han, Na Wang, Mingxi Wan Biomedical Engineering, Xi'an Jiaotong University, Xi'an, Shaanxi, China Correspondence: Meng Han Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P102 OBJECTIVES High-intensity focused ultrasound (HIFU) is a noninvasive technique for tissue ablation and tissue erosion due to the thermal or mechanical effects. However, the microstructure of the treatment target has difference after treated by the different high-intensity focused ultrasound (HIFU) therapy mode. It is important to monitor the thera- peutic effect. The Nakagami image was proposed to better visualize the tissue structure and scatter properties combined with the use of the B-modeimage simultaneously. The aim of this study was to explore the feasibility of using the Nakagami image for structure characterization of the lesions induced by focused ultrasound gradually Fig. 2 (abstract P102). The histograms of the M values of the lesion from the ablation to tissue erosion. area. (a) thermal lesion; (b) lesion by histotripsy METHODS Experiments were conducted on polyacrylamide phan- toms with bovine serum albumin (BSA) using a single-element 1.6-MHz transducer to produce tissue ablation and tissue erosion. After the treatment, ultrasound B-mode images and correspond- ing RF data was recorded and the Nakagami images was pre- P103 sented after the RFdata was processed. B-mode image and Dynamic modelling of piezoelectric hydrophones Nakagami image were combined to visualize the tissue structure Sun Yingqi, Yang Zengtao and scatter properties for structure characterization. Chongqing Medical University, Chongqing, China RESULTS For the B-mode images, the thermal lesion and the his- Correspondence: Sun Yingqi totripsy both appeared to be hyperechoic in the region of the le- Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P103 sion (Fig. 1). In the Nakagami images, the thermal lesion appeared to be homogeneous blue, and has a clear boundary OBJECTIVES The conventional hydrostatic model based on off- with the surrounding tissue. For the lesion treated by the histo- resonance hydrophone design is incapable of analyzing the tripsy, it appeared blue in most of the lesion area, and there is hydrophones operatedaroundandupon the resonance fre- an area of red in the center of the lesion. From the histogram of quencies. To expand the operating limits of the hydrostatic the M values of the lesion areas, we find that the values of the model. thermal lesion concentrated between 0.4 and 0.55 and the mean Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 116 of 122 METHODS A dynamic fluid–structure interaction model for analyz- ing the piezoelectric hydrophones is presented. An analytical so- lution for output voltage of the hydrophone operated in thickness-stretch vibration is derived. RESULTS The output voltage is proportional to the acoustic pres- sure and the piezoelectric constants. The output voltage is also inversely proportional to the elastic stiffness and dielectric con- stants of the material, and the surface area of the hydrophone. Hence, a soft material with a higher-electromechanical coupling factor will produce higher output voltage. However, around the resonance range, the output voltage is seen to strongly depend on the operating frequencies of the hydrophone and the parame- ters of the fluid. CONCLUSIONS The theoretical result shows that the output volt- age obtained by dynamic model in the off-resonance range agrees well with those obtained by hydrostatic model. Further- more, the dynamic model we present is sufficient to analyze the piezoelectric hydrophones over the entire frequency range, espe- cially around the resonance, expanding the operating limits of hydrostatic model. P104 Experimental study of micro-bubble enhance low-frequency and low-intensity ultrasound-mediated plasmid transferring into mycobacterium smegmatis Huimin Zheng Chongqing Medical University, Chongqing, China Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P104 OBJECTIVES To investigate the synergistic effect of microbubble combined with low frequency and low intensity ultrasound (LFLIU) mediated plasmid transfer into mycobacterium. METHODS A mixture of mycobacterium tuberculosis and plasmid were randomly divided into 3 groups: the control group, the ultra- sonic irradiation group and the combination of microbubble and ultrasonic irradiation group. The transformation efficiency, the vitality of the bacteria and the expression of recombinant gene ClpP2 were measured respectively by colony-counting methods, flow cytometry methods, the Real-time quantitative polymerase chain reaction after the experiment processing. RESULTS Positive colonies were selected on the LB agarose plate containing kanamycin after ultrasound exposure, the efficiency of the combination of microbubble andultrasonic irradiation group was 19.33×102CFU/ngDNA, which was more 19 times effi- cient than the ultrasonic group. The survival rate of the bacteria of ultrasonic irradiation group was 75.76%, which was more 2 times than the combination of microbubble and ultrasonic ir- radiation group. While the recombinant gene ClpP2 could be expressed, and there was no significant difference between the two groups (Fig. 1). CONCLUSIONS Ultrasound microbubble could increase the efficiency of LFLIU mediated plasmid transfer to Mycobacterium tuberculosis. The results could provide experimental reference for LFLIU-mediated Fig. 1 (abstract P104). See text for description plasmid and DNA transformation into gram positive bacteria. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 117 of 122 P105 RESULTS After 2 weeks of focused ultrasound treatment, the rate of Novel ultrasound contrast agent based on lipid containing normal in treatment group was 75.00% (15/20) and that in control sinapultide microbubbles for potential theranostics group was 10.00%(2/20),there was statistically significant between two 1,2 1,2 1,2 Dong Liu , Fang Yang , Ning Gu groups (c2=17.289,P<0.05). The total number and the degranulated Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P105 number of mast cells, the expression of PAR2, SPand NK1-R in treat- ment group were significantly lower than those in control group (all School of Biological Science & Medical Engineering, Southeast P<0.05). University, Nanjing, China; State Key Laboratory of Bioelectronics and CONCLUSIONS Focused ultrasound can treat LSC through inhibiting Jiangsu Key Laboratory forBiomaterials and Devices, Nanjing, China PAR2, SP and NK1-R expression, decreasing the total number and the Correspondence: Dong Liu degranulated number of mast cells and reducing the inflammatory Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P105 reaction of tissues. OBJECTIVES Since the bubbles were reported to have the effects of contrast enhancement, gas-filled bubbles used as ultrasonic contrast P107 agents (UCAs) have been paid great attention [1, 2]. Apart from the Feasibility study of the induction of a neuronal response in L. usage of diagnostic agents, microbubbles have also widely used as Terrestris with ultrasound stimulation 1 1 1 drug and gene delivery vehicles [3]. Accordingly, the development of Jeremy Vion , William Apoutou N'Djin , Jean-Yves Chapelon , Jahan 2,3 novel UCAs have become one of the most clinical potential fields in Tavakkoli 1 2 ultrasound medicine. In this study, we developed a novel lipid con- U1032, Inserm, Lyon, France; Physics, Ryerson University, Toronto, taining sinapultide microbubbles filled with sulfur hexafluoride (SF6). Ontario, Canada; Institute for Biomedical Engineering, Science and The good stability as well as appropriately size indicating that they Technology, Toronto, Ontario, Canada have the potential to be used as ultrasound contrast agent in Correspondence: Jeremy Vion theranostics. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P107 METHODS Stable microbubbles were fabricated by adding sinapul- tide to lipid mixture and then sonicating to generate microbubbles OBJECTIVES Several works carried out these last years have demon- by cavitation from sulfur hexafluoride (SF6) in the mixture. strated the ability of ultrasound (US) to activate neurons at a systemic RESULTS The optimized preparation of sinapultide microbubbles had level. However, no complete description of the biophysical mechanisms mean diameter of 1.82 ± 0.15μm and zeta potential of −55.2 ± 3.9 involved in this phenomenon has been validated so far. In order to mV. The acoustic imaging analysis in vitro indicated that ultrasound identify experimentally the main mechanisms responsible for ultra- imaging enhancement could be acquired by both perfusion imaging sound neurostimulation, we here present an in-vivo study of the pos- and accumulation imaging (Fig. 1). sible effects of US exposures on the generation of Compound Action CONCLUSIONS In summary, a novel lipid microbubbles encapsulated Potentials (CAPs) in the common earthworm (Lumbricus Terrestris). synthesized pulmonary surfactant sinapultide were fabricated, which METHODS We chose to study in priority the effect of US radiation makes a promising clinical potential for future theranostics of ultra- force, as proposed by Wahab et al. (2012), on a simple giant nerve sound imaging and therapy. model: the ventral nerve cord of L. terrestris. The system used was a geometrically focused piezoelectric US transducer (radius of curva- ture: 46 mm, aperture: 50 mm) working at the frequency of 0.55MHz. In each trial, a medial section of the anesthetized worm was placed within the US focal area (geometry: ellipsoidal, axial length: 29 mm at -3dB, axial width: 1.9mm at -3dB), before being exposed to US se- quences similar to those proposed by Wright et al. (2015) on the per- ipheral crab nerve model (number of cycles: 25-28 cycles/pulse, PRF = 10 kHz, number of pulses: 80-200 pulses, peak-to-peak acoustic pressure: 4-14.4 MPa). The electrophysiological response of the nerve Fig. 1 (abstract P105). See text for description was monitored with two electrodes sunken through the animal in the vicinity of one end of the nerve. Before each trial, the functional- ity of the nerve was tested by applying anelectrical stimulation (pulsed, duration = 50 μs, amplitude = 5 V, PRF = 1 Hz, pulses num- P106 ber = 1 to 10) to the other end of the nerve. Complementarily, the Effects of focused ultrasound on expression of PAR2, SP and NK1-R response of the nerve to a mechanical stimulus applied to the super- in genital skin of SD rats with vulvar lichen simplex chronicus ior end of the worm was recorded. The electrical and mechanical Huan Yang, Huajun Tang, Yijin Fan, Chengzhi Li preliminary stimulations allowed repeatable observations of the CAP College of Biomedical Engineering, Chongqing Medical University, responses, which served as a reference for studying the nerve re- Chongqing, China sponse to US exposures. Correspondence: Huan Yang RESULTS The ultrasound sequences tested allowed inducing the gen- Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P106 eration of CAPs several times. The observed CAPs presented the same characteristics as those induced by the prior electrical and OBJECTIVES To explore the expression of protease activated receptor mechanical stimulations, as shown in Fig. 1. By computing the con- 2(PAR2), substance P(SP) and neurokinin-receptors in the genital skin duction velocity associated to the observed CAPs, we identifiedthe of SD rats with vulvar lichen simplex chronicus(LSC) after focused nervous structures stimulated to be the Medial Giant Fiber (MGF) ultrasound treatment. and the Lateral Giant Fiber (LGF). Over the trials, both fibers were METHODS Forty female rat models of LSC were established by stimulated at least once butnever simultaneously. However, a great Sally method, and then the rats were randomly divided into number of trials were performed, using various level of acoustic pres- treatment group (n=20) and control group(n=20). The rats in the sure, and this phenomenon appeared to be very littlerepeatable. treatment group were treated by focused ultrasound, and those CONCLUSIONS Several hypotheses were considered to explain the low in the control group received sham treatment. After focused repeatability of our US neurostimulation results. First, it may be neces- ultrasound intervention, the histological changes of vulva were sary to target specific areas in the giant nerve such as dorsal nodes, observed by using HE staining. The total number and the degra- which could be performed in our future trials by using ultrasound guid- nulated number of mast cells were identified by staining slides ance. Second, we cannot be sure that uncontrolled episodes of cavita- with toluidine blue. Immunohistochemistry was used to test the tion did not arise during some trials. This hypothesis is currently being expression of PAR2, SP and NK1-R. Differences among the groups tested in a series of trials where the ultrasound exposure is compatible were compared. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 118 of 122 with the controlled generation of cavitation. Finally, the myelin sheath RESULTS Ten specimens of PDA were scanned using HMI (5 from dis- embedding the giant nerve could also have an impact on the possibil- tal pancreatectomy and 5 from Whipple procedure). Preliminary hist- ity of inducing mechanical effects on the neural membrane. Hence, we ology results showed that the scan did not damage the tissue in any plan to adapt our set-up to the ventral nerve cord of the american lob- aspect, and did not impact the subsequent analysis. HMI was able to ster (Homarus Americanus). This unmyelinated nervous model has scan the full volume of the specimens with measurements asdeep as already been the object of preliminary studies performed during an 3 cm in the tissue. The whole range of stiffness from tumor to nor- international collaboration. In conclusion, we were able to demonstrate mal pancreas was assessed. The Fig. 1 shows the B-mode image of with this study the feasibility of inducing nervous responses with US the maximal cross-section of the tumor, and the HMI map overlaid and these responses were identical to those generated with electrical on the B-mode image. The mean displacement inside the tumor is and mechanical stimulation. Further investigations are necessary to 1.41 ± 1.00 μm versus 11.25 ± 3.69 μm in the surrounding tissue. control this phenomenon and make it more repeatable. This project Measurements in normal pancreatic tissue demonstrated a mean was supported by the Laboratory of Excellence (LabEx) DevWeCan and HMI displacement of 17.05 ± 3.98μm. A sharp mechanical contrast is the MitacsGlobalink Research Award - Campus France. evident on the HMI maps between the tumor and the surrounding tissue which correlates very well with the B-mode image. CONCLUSIONS This initial feasibility study showed that HMI is cap- able of scanning whole resected pancreatic cancer specimens with- out any damage to the tissue. The technique can assess deep parts of the tissue with greatly difference stiffness in a limited amount of time. HMI significantly detected PDA within the specimen with mean displacement lower by a ratio of 9.8 compared to the surrounding tissue and 12.1 compared to normal pancreas away from the mass. This indicates the possible presence of fibrotic tissue close to the tumor. The sharp tumor delineation observed on HMI maps was con- firmed on the B-mode image.HMI is thus shown promising for pan- creatic cancer detection and characterization. The technique can provide images of the entire organ and preliminary results indicate that it can distinguish PDA, fibrotic tissue from normal pancreatic tis- sue. HMI could thus constitute a very important tool for screening patients at risk of developing pancreatic cancer and showing the non-specific symptoms. Fig. 1 (abstract P107). Examples of nervous responses to an electrical stimulus (a), a mechanical stimulus (b) and an ulstrasound stimulus (c) P108 Fig. 1 (abstract P108). Pancreatic ductal adenocarcinoma in a distal Harmonic Motion Imaging (HMI) for pancreatic cancer detection pancreatectomy specimen. The Bmode image (top) is shown as well and characterization in post-surgical human specimens as the overlaid HMI displacement map (bottom) 1 2 1,3 Thomas Payen , Kenneth Olive , Elisa E. Konofagou Biomedical Engineering, Columbia University, New York, New York, USA; Medicine and Pathology & Cell Biology, Columbia University Medical Center, New York, New York, USA; Radiology, Columbia University Medical Center, New York, New York, USA Correspondence: Thomas Payen P109 Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P108 Experimental study on the influence of low-frequency and low- intensity ultrasound on the permeability of the mycobacterium OBJECTIVES Pancreatic ductal adenocarcinoma (PDA) has one of the smegmatis cytoderm and potentiation with levofloxacin 2 3 2 1 lowest prognosis due to non-specific symptoms leading to late diag- Yu Dong , Hang Su , Yonghong Du , Dairong Li nosis and therapeutic inefficiency. PDA is characterized by an un- Department of Respiratory disease, The First Affiliated Hospital of usually dense stroma limiting chemotherapy perfusion. Harmonic Chongqing Medical University, Chongqing, China; Chongqing medical Motion Imaging (HMI) assesses tissue mechanical properties by indu- university, Chongqing, China; Food and Drug Administration of Huiji, cing localized oscillation resulting from a periodic acoustic radiation Zhengzhou, China force. The amplitude of the induced displacement is directly related Correspondence: Yu Dong to the underlying tissue stiffness. The objective of this study was to Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P109 evaluate the utility of HMI in post-surgical human pancreatic cancer OBJECTIVES Tuberculosis (TB) is an infectious disease caused by the specimens and its capability of differentiating the tumor for perile- sional fibrotic tissue and normal tissue. bacterium Mycobacterium tuberculosis. The aim of this study is to in- vestigate the synergistic potentiating effect of low-frequency and METHODS A 4.5-MHz focused ultrasound transducer (FUS) generates an low-intensity ultrasound (LFLIUS) with levofloxacin on M. smegmatis amplitude-modulated beam resulting in harmonic tissue oscillations at its (MS) [a "surrogate of MTB"] and to explore underlying mechanisms. focus. Axial tissue displacement is estimated using 1D cross-correlation of RF signals acquired with a 2.5-MHz diagnostic transducer (P4-2, ATL) METHODS M. smegmatis was continuously irradiated by ultrasound using a plane-wave beam sequence, confocally aligned with the FUS.Ten of 42 kHz with several different doses (intensity and duration), human pancreatic specimens (approximate dimensions = 24 x 12 x 80 respectively. The bacteria vitality, structure and morphology were mm3) were assessed immediately after surgery. They were immersed in subjected to plate counting and microscopic examinations. Flow cy- PBS and their full volume was scanned for a maximum duration of 90 mi- tometry was adopted to test the fluorescence intensity of bacteria nutes. The imaging planes were chosen perpendicular to the pancreatic that were stained by FAD, or PI dye. Spatially and temporally aver- duct to correlate the HMI displacement withstandard histology analysis. aged intensity (ISATA) of 0.138 W/cm was then acted on Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 119 of 122 mycobacterium smegmatis combined with levofloxacin to confirm P111 the synergy between ultrasound and antibiotic. The safety on subsequent pregnancy in women after ultrasound RESULTS The results showed that the permeability of M. smegmatis ablation of uterine fibroids: a single-central retrospective study increased after ultrasound exposure; and the survival rate, structure Junshu Li, Wenzhi Chen, Jinyun Chen and morphology of bacteria of the low-dose (ISATA= 0.138 W/cm The State Key Laboratory of Ultrasound Engineering in Medicine Co- for 5 min) treated ultrasound group had no evident difference Founded by Chongqing and the Ministry of Science and Technology, compared with those of the control group (P>0.05) ; but the survival Chongqing Key Laboratory ofBiomedical Engineering, College of rate of bacteria significantly decreased and the bacteria structure got Biomedical Engineering, Chongqing Medical University, Chongqing serious damage in high-dose (ISATA= 0.329 W/cm for 20 min) Collaborative Innovation Center for Minimally-invasive andNoninvasive ultrasound group. The bactericidal effect of ultrasound combining Medicine, ChongQing, China with levofloxacin was evidently higher than that of single ultrasound Correspondence: Junshu Li or single levofloxacin. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P111 CONCLUSIONS A certain dose of LFLIUS can increase the permeabil- ity of M. smegmatis’ cytoderm and exert a synergy to bactericidal ac- OBJECTIVES A retrospective analysis to explore the impact of high- tion of levofloxacin. intensity focused ultrasound (HIFU) ablation of uterine fibroids in women on subsequent pregnancy. METHODS A retrospective analysis was conducted of women with uterine fibroids who underwent HIFU ablation at the Clinical Center for Tumor Therapy, Chongqing Medical University, Chongqing, China, P110 from January 1, 2010, to January 1, 2015. Inclusion criteria were: (1) Intracellular calcium response of hela cells stimulated by the aged 20-42 years (2) with fertility desire (3) have normal sexual life impulsive jetting flow from tandem bubble interaction without contraception after HIFU. Exclusion criteria were: fertility bar- 1 1 2 1 3 Fenfang Li , Chen Yang , Fang Yuan , Defei Liao , Guilak Farshid , Pei riers unrelated to HIFU, such as hysterectomy or bilateral oophorec- Zhong tomy. The registry of patient cases were screened by marital status Mechanical Engineering and Materials Science, Duke University, and birth history before treatment, and telephone follow-up informa- Durham, North Carolina, USA; HuaCells Corporation, Natick, tion including improvement of symptoms of uterine fibroids, sexual Massachusetts, USA; Department of Orthopaedic Surgery, Washington life, pregnancy, and delivery information after HIFU treatment. University, St. Louis, Missouri, USA RESULTS A total of 189 cases met inclusion criteria and followed-up ef- Correspondence: Fenfang Li fectively, the median follow-up time was three years. Majority of them Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P110 have symptoms of uterine fibroids improve significantly after HIFU treat- ment, and volume of uterine fibroids decrease(P<0.05). Among the 189 OBJECTIVES Cavitation plays an important role in the bioeffects pro- cases, there were 131 cases pregnancy with a total of 133 times.Of 131 duced by therapeutic ultrasound. However, limited studies have been pregnant women,19 were on-going pregnancy, 21 cases of abortion, the carried out to investigate the mechanisms of cellular mechanotransduc- remaining 91 cases in successful pregnancies and deliveries.The inci- tion that are activated by bubble pulsations in a controlled manner, espe- dence rates of complications during pregnancy and labor were 10.8% cially at the single cell level. Mechanotransduction is a process through (10/93) and 7.5% (7/93), respectively.The most common complications of which cells sense and respond to mechanical stimuli (e.g., membrane pregnancy in this study were placenta previa and placenta implant- tension and strain) and convert them to biochemical signals. In many ation(5/10).The incidence rate of massive hemorrhage during labor was cases, the initial mechanotransduction signal involves an increase in intra- 6.5%(6/93),the majority (66.7%) were associated with resection intramural cellular calcium ion (Ca2+) elicited by mechanical stress, which subse- fibroid during cesarean section.The study presented here did not contain quently triggers various downstream biochemical reactions that may any cases of uterine rupture during pregnancy or labor after HIFUtreat- alter the morphology and function of the cell. In this study, we examine ment. Ninety-one women successfully delivered 93 times with a cesarean the intracellular calcium response of single adherent HeLa cells evoked section rate of 72.0%(67/93). Among them,only fourteen women (20.9%) by the directional jetting flow from tandem bubbles (Rmax = 50 ± 2 μm). chose delivery by cesarean section for obstetric factors, while 79.1% METHODS Bubble dynamics, morphology, location, and adhesion chose delivery by cesarean section for social factors. conditions of individual cells were well-controlled in a microfluidic CONCLUSIONS In conclusion, HIFU is a safe and effective noninvasive system with surface patterning. Two types of calcium responses, in therapy to treat uterine fibroids in women who wish to retain the ability the form of intracellular calcium waves (ICW), were observed at dif- to conceive and deliver after treatment. Additionally, HIFU effectively pro- ferent standoff distance (Sd = 30 to 60 μm) of the tandem bubbleto vides remission of symptoms, has a low rate of complications through the cell through time-elapsed fluorescence imaging. pregnancy and labor, and does not increase the rate of spontaneous RESULTS The ICW was initiated from the leading edge facing the jet- abortion or delivery by cesarean section. Based on our findings presented ting flow, and propagated toward the trailing edge of the cell. At Sd/ here, we believe that HIFU should be recommended treatment for uter- Rmax = 0.6, a fast calcium response with short rise time and large amp- ine fibroids in women who wish to retain the ability to have children in litude in calcium concentration change was produced. In contrast, at the future, or who otherwise wish to not undergo hysterectomy. Sd/Rmax = 1, a slow calcium response with long risetime and small amplitude was often observed. In all cases, the calcium response was initiated by the influx of extracellular calcium through the membrane by either poration produced predominately at Sd/Rmax = 0.6 or ion P112 channel opening induced primarily at Sd/Rmax = 1.0, which were con- Low intensity pulsed ultrasound relieves myelosuppression firmed by mechanistic studies using calcium-free extracellular medium, induced by cyclophosphamide in rabbits membrane poration indicator propidium iodide, or the non-specific B. Liu mechanosensitive ion channel blocker Gd3+. The elicited calcium tran- Chongqing Medical University, Chongqing, China sient propagated in the cytosol through calcium-induced calcium re- Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P112 lease (CICR) from endoplasmic reticulum storage, which was inhibited by thapsigargin. We further demonstrated that attaching 6 μm integ- OBJECTIVES This study is to investigate the effect of low intensity rin-binding beads to the cell membrane can trigger calcium response pulsed ultrasound (LIPUS) on cyclophosphamide (CTX)-induced rabbit under conditions (Sd/Rmax = 1.2) that do not elicit such a response by myelosuppression. the tandem bubbles without the beads. METHODS A total of 40 New Zealand white rabbits were used to es- CONCLUSIONS These findings show that directional jetting flow from tablish the myelosuppression models by daily ear vein injection of 40 tandem bubbles can be used to induce highly controlled mechanical mg/kg CTX for 4 continuous days. They were randomly divided into stimulation on individual cells. LIPUS group and control group. LIPUS was performed once a day for Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 120 of 122 20 minutes each time for 7 and 28 days, respectively. The treatment METHODS Targestar-SA microbubbles conjugated with RGD li- effects of LIPUS were examined by comparing general conditions, gands wereusedto attach on thecell membraneofHeLa cells. mortality rates, blood routine and bone marrow hyperplasia in bone A Plasmid encoding for green fluorescent protein (GFP) was used marrow smears. The safety of the LIPUS irradiation was evaluated by for gene transfection. Two ultrasound conditions were selected to HE staining. stimulate two typical bubble activities while achieving good RESULTS LIPUS improved the number of peripheral blood cells and transfection efficiency and maintaining high cell viability, high bone marrow nucleated cells and thus reduced the mortality of pressure short pulse (0.6Mpa, 10μs) and low pressure long pulse model rabbits in different degrees. The irradiated parts of the skin (0.2MPa, 1ms). Multiply approaches were employed to investigate did not burn, and the muscle tissue in acoustic channels showed nor- bubble dynamics, cell membrane responses and plasmid DNA in- mal structures and no obvious pathological changes, suggesting ternalization routes: (i) Fast frame video microsocopy (Photron LIPUS irradiation is safe for the model rabbits. FASTCAM SA2) was used to record the microbubble dynamics. (ii) CONCLUSIONS LIPUS is a safe and effective method to relieve CTX- Real-time fluorescent microscopy of propidium iodide to indicate induced myelosuppression, which can be used for the prevention and and characterize pores. (iii) Confocal fluorescent microscopy of treatment of the occurrence and development of myelosuppression. plasmid DNA (Cy3 labled), cell membrane (Alexa Fluor 647 labled), and nuclear (Hoechst 33342 labeled) (Fig. 1). And (iv) SEM images for the surface topography of cell membrane. All P113 these observations were spatiotemporally correlated. Cost-effectiveness evaluated by markov decision-making tree RESULTS (i) When stimulated by a high pressure short pulse model of high-intensity focused ultrasound ablation versus (0.6Mpa, 10μs), targeted microbubble generates a transient and surgery for uterine fibroids reversible pore on the cell membrane, with radius varies from Liang Hu, Jinyun Chen, Yujie Feng, Chang Liu, Wenzhi Chen, Zhibiao several microns to several hundreds microns. While driven by low Wang pressure long pulse (0.2MPa, 1ms), the microbubbles undergo College of Biomedical Engineering, Chongqing Medical University, gentle oscillation, unlikely to perforate cell membrane. (ii) When Chongqing, China driven by high pressure short pulse, plasmid DNA can be directly Correspondence: Liang Hu and promptly propelled into the cytosol, even nuclear, through Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P113 pores with the diameter larger than about 100 microns. (iii) When driven by low pressure long pulse, the gentle bubble-cell inter- OBJECTIVES To investigate the cost-effectiveness of high-intensity fo- action significantly enhanced the plasmid DNA aggregation and cused ultrasound (HIFU) ablation versus surgery (myomectomy and retention onto the cell membrane, while the global endocytosis hysterctomy) for uterinefibroids. is responsible for plasmid DNA internalization. (iv) The determin- METHODS Direct medical cost, indirect cost, efficacy and safety out- ant factor for different plasmid DNA uptake processes is the comes of patients with uterine fibroids who ever underwent HIFU, ultrasound driven bubble- cell interaction. myomectomy, and hysterectomywere collected to build Markov Deci- CONCLUSIONS By establishing direct spatiotemporal correlation sion-making Tree Model based on health status of these patients. between ultrasound-stimulated targeted microbubbles activities The Model was employed to make simulation of health status transi- and the resulting intracellular plasmid DNA uptake process, this tion for uterine fibroids patients treated by these different proce- study revealed the complex and bubble-cell interaction dures. Cost-effectiveness, cost-utility as well as sensitivity analysis dependent nature of plasmid DNA uptake mechanisms at sin- were conducted for evaluation ofHIFU, myomectomy and hysterec- gle cell level. tomy for uterine fibroids. RESULTS The Markov Decision-making Tree Model simulation showed that each unit improvement of physical summary scale (PCS) and mental summary scale (MCS)related to HIFU ablation consumed 1782.16 and 1120.88 respectively. By contrast, each unit improve- ment of PCS and MCS related to surgery consumed 1825.43 and 1605.15 respectively. CONCLUSIONS The Markov Decision-making Tree Model simulation and healthy economic analysis demonstrate HIFU is dominatingly cost-effective for uterinefibroids. P114 Mechanisms of plasmid DNA intracellular uptake facilitated by ultrasound and targeted microbubbles is determined by bubble-cell interaction 1 2 2 1 N. Rong , H. Zhou , R. Liu , Z. FAN Biomedical Engineering, Tianjin University, Tianjin, Tianjin, China; College of Life Sciences, Nankai University, Tianjin, Tianjin, China Correspondence: N. Rong Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P114 OBJECTIVES Therapeutic Ultrasound combined with microbubbles can achieve non-viral, targeted delivery of genetic materials, as a promising delivery technique for gene therapy. However, insuffi- cient understanding of plasmid DNA intracellular uptake process Fig. 1 (abstract P114). Cy3 labeled plasmid DNA distribution in and the important role of microbubble plays during this process, Hela cells of different ultrasound parameter. Bright field images limits the progress of improving gene transfection efficiency and showing bubble information and representative confocal images translating this technique into clinics. Therefore, the aim of this showing pDNA distribution with Cy3 labeled pDNA (red), cell nuclei study is to investigate the intracellular uptake mechanisms of (blue) and membrane(yellow).(A) 0.6MPa,10μs, (B) 0.2MPa,1ms,(C) plasmid DNA mediated by targeted microbubbles driven by dif- Control without ultrasound stimulation.Scale bar is 5um ferent ultrasound conditions. Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 121 of 122 P115 doses of 0 – 20 Gy. The analysis was conducted by using three differ- Optimization of tumour therapy by combination of focused ent cell lines for prostate cancer (PC-3, Vcap, LNcap), glioblastoma ultrasound and radiation guided by MRI and PET-MRI (LN405, U87, T98G) and head/neck tumor (FaDu, UT-SCC 5, UT-SCC 1 1 1 1 2 Doudou Xu , Lisa Landgraf , Xinrui Zhang , Michael Unger , Ina Patties , 8). Effects at the cellular level on metabolism (WST-1), proliferation 1 1 3,4 3,5 Johann Berger , Shaonan Hu , Lydia Koi , Antje Dietrich , Aswin (BrdU), membrane integrity (LDH release) and apoptosis (Annexin V) 3,4 3,4 1 1 Hoffmann , Mechthild Krause , Thomas Neumuth , Andreas Melzer were detected after treatment. Innovation Center Computer Assisted Surgery, Universität Leipzig, In vivo: PET-MR and MR guided focused ultrasound system allows Leipzig, Germany; Department of Radiation Therapy, Universität Leipzig, precise sonication treatment for small animals bearing tumors, under Leipzig, Germany; OncoRay - National Center for Radiation Research in real-time MR-thermometry [6]. Small animal PET-MR system (nanoS- Oncology, Dresden, Germany; Department of Radiation Oncology, can, Mediso) will be employed and integrated with a FUS transducer University Hospital Carl Gustav Carus, Technische Universität Dresden, (11×11 matrix array), which allows the function of beam forming to Dresden, Germany; German Cancer Consortium (DKTK), partner site achieve hyperthermia treatment. Local tumor irradiations under nor- Dresden, German Cancer Research Center (DKFZ), Heidelberg, Germany mal blood flow conditions will then be given with 200 kV X-rays (0.5 Correspondence: Doudou Xu mm copper -filter) and 20 mA at a dose rate of ~ 1.1 Gy/min (X-ray Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P115 machine type Yxlon Y.TU 320-D03) [7]. The self-filtering of the X-ray tube is 3 mm beryllium and 3 mm aluminum. Nude mice bearing Introduction heterotopic tumors on hind leg will be treated by both FUS and RT The two ZIK-Centers for Innovation Competence, ICCAS in Leipzig in sequence. Tumor size monitoring, histology study and UPLC mo- and OncoRay in Dresden, have joined forces to start a new multidis- lecular analysis will be investigated post the combination therapy. ciplinary 6.3 million Euro research project: SONO-RAY - Tumor ther- Robot installed in PET-MR: A MR-compatible robotic arm system TM apy combining image-guided (PET-MR and MR) focused ultrasound (INNOMOTION , Innomedic) [8, 9] was installed with Biograph mMR and radiation therapy. The goal of SONO-RAY is to combine noninva- MR-PET (Siemens Healthineers) in the Department of Nuclear Medi- sive image-guided therapy approaches of magnetic resonance cine of the University Medical Center Leipzig to investigate the ef- guided focused ultrasound and radiation therapy to improve the effi- fects of a combination of focused ultrasound and radiation therapy. cacy of cancer treatment. The robotic arm will reposition the ultrasound transducer during the sonication treatment. It is possible to detect residual tumor tissue Hypothesis and Motivations: after the treatment by using PET-MR imaging to provide an optimal SONO-RAY project aims to combine therapeutic focused ultrasound treatment outcome. (FUS) and radiation therapy (RT) to treat malignant solid tumors and MR guided FUS-Sonalleve: MR-guided high-intensity focused ultra- tumor metastases. The hypothesis underlying this approach is that sound (MR-HIFU) is a noninvasive technique for depositing thermal the combination of two tissue-destroying energies (the energy of energy in a controlled manner deep within the body. The Philips high-intensity acoustic waves and ionizing radiation) is more effect- Sonalleve MR-HIFU system, initially released in 2010, was installed in ive in cancer treatment than the effect of employing one of the Leipzig University Hospital at the beginning of 2017, introduced a above two energy forms alone [1-3]. The central scope of the project new approach for uterine fibroids[10, 11] and bone metastasis treat- is to develop tumor cell biology fundamentals, the computer-aided ment [12, 13] by employing thermal ablation. Sonalleve system also model formation and to evaluate the success potential of a future offers solutions of hyperthermia research platform [14, 15] in combin- clinical use of a FUS-RT combination therapy. Within the framework ation with radiation therapy and chemotherapy in cancer treatment. of this project, the basic principle of the combined FUS and RT effect MR guided FUS-prostate system: The TULSA-PRO System (Profound on tumor cells (in vitro) and the tumor tissue (in vivo) is going to be Medical) is a transurethral MR guided FUS system for whole gland investigated preclinically. Clinical applications could be the treatment ablation of the prostate [16]. A test system was installed at the uni- of various tumors of the head and neck, prostate carcinoma and versity hospital Dresden to perform the world’s first compatibility glioblastoma. tests on a Philips Ingenuity TF PET/MR scanner. The system com- Project work package: prises a transurethral ultrasound applicator with 10 FUS elements In order to provide the basis for the combined FUS-RT method, in vitro working at 4 or 14 MHz, depending on the penetration depth re- and in vivo experiments are being be carried out to elaborate the ther- quired to heat the rim of the prostate (i.e. the so-called “thermal ab- mal and mechanical effect on the tumor tissue [4]. After identification lation boundary”) up to 55 deg Celcius. The system has an of the thermal and mechanical individual parameters for FUS, these are endorectal cooling device to prevent the rectal mucosa from over- transferred into simulation models in order to predict the behavior of heating. The ultrasound applicator also has a cooling circuit to spare the fabric on FUS[5]. The simulation models are validated using the pa- the urethra. The power and frequency of the 10 ultrasound transduc- rameters. Based on the data obtained above, FUS and RT will be gener- ers are individually steered by real-time MR-thermometry. ated spatially dispersed biologically active doses. Algorithms will be used for the anatomical co-registration and accumulation of the bio- logically active dose distributions generated by FUS and RT on MR im- References ages. Furthermore, software modules will be developed and validated, [1] R. Cirincione, F.M. Di Maggio, G.I. Forte, L. Minafra, V. Bravata, L. Castiglia, which support the clinical user in therapy planning. Optimization of ap- V. Cavalieri, G. Borasi, G. Russo, D. Lio, C. Messa, M.C. Gilardi, F.P. plication sequences of the FUS-RT technology and the incorporation in Cammarata, High-Intensity Focused Ultrasound- and Radiation Therapy- therapy into the treatment process (clinical workflow) of the patient will Induced Immuno-Modulation: Comparison and Potential Opportunities, also be investigated and developed. Ultrasound in medicine & biology, 43 (2017) 398-411. Experimental facilities in SonoRay [2] T. Refaat, S. Sachdev, V. Sathiaseelan, I. Helenowski, S. Abdelmoneim, M.C. In vitro FUS and RT: A high throughput in vitro 96-well sonicator Pierce, G. Woloschak, W. Small, Jr., B. Mittal, K.D. Kiel, Hyperthermia and was designed and allows individual sonication for each of the wells radiation therapy for locally advanced or recurrent breast cancer, Breast, in a 96-well plate. It consists of 96 single transducers at an operating 24 (2015) 418-425. frequency of 1 MHz and a maximum energy of 0.05 W/cm . A 150 kV [3] S.K.Wu, C.F.Chiang,Y.H. Hsu, H.C. Liou, W.M. Fu,W.L.Lin, X-ray machine (DARPAC 150-MC) was employed for irradiation at Pulsed-wave low-dose ultrasound hyperthermia selectively enhances Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):2 Page 122 of 122 nanodrug delivery and improves antitumor efficacy for brain P116 metastasis of breast cancer, Ultrasonics sonochemistry, 36 Combination therapy for cancer by PET-MR and MR image guided (2017) 198-205. focused ultrasound and radiation 1 1 1 1 2 [4] J. Beik, Z. Abed, F.S. Ghoreishi, S. Hosseini-Nami, S. Mehrzadi, A. Shakeri- Doudou Xu , Lisa Landgraf , Xinrui Zhang , Michael Unger , Ina Patties , 1 1 3,4 5 Zadeh, S.K. Kamrava, Nanotechnology in hyperthermia cancer therapy: Johann Berger , Shaonan Hu , Lydia Koi , Marc Fournelle , Steffen 5 1 1 From fundamental principles to advanced applications, Journal of con- Tretbar , Thomas Neumuth , Andreas Melzer trolled release : official journal of the Controlled Release Society, 235 Innovation Center Computer Assisted Surgery, Universität Leipzig, (2016) 205-221. Leipzig, Germany; Department of Radiation Therapy, Universität [5] J. Hu, Y. Ding, S. Qian, X. Tang, Simulations of adaptive temperature Leipzig, Leipzig, Germany; OncoRay - National Center for Radiation control with self-focused hyperthermia system for tumor treatment, Ul- Research in Oncology, Dresden, Germany; Department of Radiation trasonics, 53 (2013) 171-177. Oncology, University Hospital Carl Gustav Carus, Technische Universität [6] A.J. Loeve, J. Al-Issawi, F. Fernandez-Gutierrez, T. Lango, J. Strehlow, S. Dresden, Dresden, Germany; Fraunhofer IBMT, St. Ingbert, Germany Haase, M. Matzko, A. Napoli, A. Melzer, J. Dankelman, Workflow and inter- Correspondence: Doudou Xu vention times of MR-guided focused ultrasound - Predicting the impact Journal of Therapeutic Ultrasound 2018, 6(Suppl 1):P116 of new techniques, Journal of biomedical informatics, 60 (2016) 38-48. [7] A. Yaromina, T. Kroeber, A. Meinzer, S. Boeke, H. Thames, M. Baumann, D. INTRODUCTION FUS-RT stands for the combination therapy of fo- Zips, Exploratory study of the prognostic value of microenvironmental cused ultrasound (FUS) and radiation therapy (RT). The aim of parameters during fractionated irradiation in human squamous cell SONORAY project is to use the synergistic effect of heat and carcinoma xenografts, International journal of radiation oncology, mechanical effects of sound as well as ionizing radiation in order biology, physics, 80 (2011) 1205-1213. to improve the results of radiooncological treatments of malig- [8] A.J. Krafft, J.W. Jenne, F. Maier, R.J. Stafford, P.E. Huber, W. Semmler, M. nant solid tumors and metastases. Physical and biological effects Bock, A long arm for ultrasound: a combined robotic focused ultrasound are being investigated and the synergy of FUS and RT will be setup for magnetic resonance-guided focused ultrasound surgery, Med- quantified in simulation, cell, and small animal studies. The work ical physics, 37 (2010) 2380-2393. is accompanied by the development of multimodal treatment [9] N.V. Tsekos, A. Khanicheh, E. Christoforou, C. Mavroidis, Magnetic planning and information systems to prepare for a seamless inte- resonance-compatible robotic and mechatronics systems for image- gration into the clinical workflow. The project aims at developing guided interventions and rehabilitation: a review study, Annual review of a proof-of-concept system and workflow for the translation into biomedical engineering, 9 (2007) 351-387. clinical use. [10] Y.S. Kim, B. Keserci, A. Partanen, H. Rhim, H.K. Lim, M.J. Park, M.O. Kohler, METHODS A high throughput in vitro sonicator with 1.14 MHz single Volumetric MR-HIFU ablation of uterine fibroids: role of treatment cell transducer made by piezoelectric ceramic material was employed size in the improvement of energy efficiency, European journal of radi- and allows individual sonication for wells in a 96-well plate. T98G gli- ology, 81 (2012) 3652-3659. oma cells were exposure to acoustic intensity of 71 W/cm with [11]M.J. Voogt, H. Trillaud, Y.S. Kim, W.P. Mali, J. Barkhausen, hyperthermia (40-45°C) duration of 134 sec, and 109 W/cm with L.W. Bartels, R. Deckers, N. Frulio, H. Rhim, H.K. Lim, T. Eckey, hyperthermia (40-45°C) duration of 65 sec, respectively. A 150 kV X- H.J. Nieminen, C. Mougenot, B. Keserci, J. Soini, T. Vaara, M.O. ray machine (DARPAC 150-MC) was employed for irradiation at doses Kohler, S. Sokka, M.A. van den Bosch, Volumetric feedback of 0-20 Gy to investigate the radiation curve of T98G cells. For FUS ablation of uterine fibroids using magnetic resonance-guided and RT combinations, 5 and 10 Gy were used to treat T98G cells high intensity focused ultrasound therapy, European radiology, 24hrs post sonication. Effects at the cellular level on metabolism 22 (2012) 411-417. (WST-1), proliferation (BrdU) and membrane integrity (LDH release) [12]M. Huisman, M.K. Lam, L.W. Bartels, R.J. Nijenhuis, C.T. Moonen, were detected after treatment. F.M. Knuttel, H.M. Verkooijen, M. van Vulpen, M.A. van den Bosch, RESULTS The preliminary RT results showed dose dependent loss in Feasibility of volumetric MRI-guided high intensity focused ultrasound cellular NAD(P)H levels of 60% for T98G cell lines at 20 Gy. A release (MR-HIFU) for painful bone metastases, Journal of therapeutic of LDH was observed from 4% (0 Gy) to 17% (20 Gy). The highest im- ultrasound, 2 (2014) 16. pact of RT was detected during analysis of DNA synthesis (BrdU) [13]M.K. Lam, M. Huisman, R.J. Nijenhuis, M.A. van den Bosch, which nearly stopped at dosages above 5 Gy. In terms of the first M.A. Viergever, C.T. Moonen, L.W. Bartels, Quality of MR thermometry FUS and RT combination experiment, there is no higher LDH release during palliative MR-guided high-intensity focused ultrasound in the combination therapy group in comparison to FUS or RT single (MR-HIFU) treatment of bone metastases, Journal of therapeutic treatment groups. In contrast, cells treated by combination of FUS ultrasound, 3 (2015) 5. hyperthermia at 40-45°C for 65 secs with 10 Gy irradiation treatment [14]E.J. Dorenberg, F. Courivaud, E. Ring, K. Hald, J.A. Jakobsen, E. suggested lower cell viability of 81% (WST-1 assay) in comparison to Fosse, P.K. Hol, Volumetric ablation of uterine fibroids using treatment of FUS hyperthermia only (97% viable cells) and RT only Sonalleve high-intensity focused ultrasound in a 3 Tesla scanner– (90% viable cells). first clinical assessment, Minimally invasive therapy & allied tech- CONCLUSIONS In conclusion, longer FUS hyperthermia duration nologies : MITAT : official journal of the Society for Minimally Inva- from 100 sec to 1000 sec should be further investigated in sive Therapy, 22 (2013) 73-79. combination with RT treatment. The interval between FUS [15] N.M. Hijnen, E. Heijman, M.O. Kohler, M. Ylihautala, G.J. Ehnholm, A.W. hyperthermia and irradiation treatment will be conducted within Simonetti, H. Grull, Tumour hyperthermia and ablation in rats using a 8hrs according to literatures. A fitted treatment for different cell clinical MR-HIFU system equipped with a dedicated small animal set-up, lines will be necessary based on the different radiosensitivities. International journal of hyperthermia : the official journal of European So- Future in vitro investigations of effects of FUS hyperthermia as ciety for Hyperthermic Oncology, North American Hyperthermia Group, well as of a combined therapy on other tumor cells need to be 28 (2012) 141-155. conducted. [16] M. Burtnyk, T. Hill, H. Cadieux-Pitre, I. Welch, Magnetic resonance image guided transurethral ultrasound prostate ablation: a preclinical safety and Publisher’s Note feasibility study with 28-day followup, The Journal of urology, 193 (2015) Springer Nature remains neutral with regard to jurisdictional claims in published 1669-1675. maps and institutional affiliations.

Journal

Journal of Therapeutic UltrasoundSpringer Journals

Published: May 15, 2018

There are no references for this article.