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

Learn More →

Respiratory level tracking with visual biofeedback for consistent breath-hold level with potential application in image-guided interventions

Respiratory level tracking with visual biofeedback for consistent breath-hold level with... Background: To present and evaluate a new respiratory level biofeedback system that aids the patient to return to a consistent breath-hold level with potential application in image-guided interventions. Methods: The study was approved by the local ethics committee and written informed consent was waived. Respiratory motion was recorded in eight healthy volunteers in the supine and prone positions, using a depth camera that measures the mean distance to thorax, abdomen and back. Volunteers were provided with real-time visual biofeedback on a screen, as a ball moving up and down with respiratory motion. For validation purposes, a conversion factor from mean distance (in mm) to relative lung volume (in mL) was determined using spirometry. Subsequently, without spirometry, volunteers were given breathing instructions and were asked to return to their initial breath-hold level at expiration ten times, in both positions, with and without visual biofeedback. For both positions, the median and interquartile range (IQR) of the absolute error in lung volume from initial breath-hold were determined with and without biofeedback and compared using Wilcoxon signed rank tests. Results: Without visual biofeedback, the median difference from initial breath-hold was 124.6 mL (IQR 55.7–259.7 mL) for the supine position and 156.3 mL (IQR 90.9–334.7 mL) for the prone position. With the biofeedback, the difference was significantly decreased to 32.7 mL (IQR 12.8–59.6 mL) (p < 0.001) and 22.3 mL (IQR 7.7–47.0 mL) (p <0.001), respectively. Conclusions: The use of a depth camera to provide visual biofeedback increased the reproducibility of breath-hold expiration level in healthy volunteers, with a potential to eliminate targeting errors caused by respiratory movement during lung image-guided procedures. Keywords: Breath holding, Image-guided biopsy, Lung, Reproducibilty of results, Respiration Key points  This method can have potential application in lung image-guided interventional procedures, reducing A depth camera can be used to accurately monitor targeting errors caused by respiration the level of respiration Visual feedback enables volunteers to hold their Background breath at a consistent level Tumour movement caused by patient respiration can be a serious problem during image-guided interventional procedures. Lung nodules near the diaphragm, for * Correspondence: w.j.heerink@umcg.nl Center for Medical Imaging – North East Netherlands, University of example, typically move > 2 cm with inspiratory capacity Groningen, Groningen, The Netherlands [1]. When tissue diagnosis of a suspicious lung nodule Department of Radiology, University of Groningen, University Medical by computed tomography (CT)-guided biopsy is Center Groningen, Groningen, The Netherlands Full list of author information is available at the end of the article required, a predictable nodule position is important. © 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. Heerink et al. European Radiology Experimental (2018) 2:22 Page 2 of 9 Often, giving a patient specific breathing instructions is was angled forward towards the patient, so the patient not sufficient to facilitate a consistent breath-hold level. remained in its field of view, while in the CT gantry. For When patients have difficulty returning to their initial convenience and lack of continuous access to a CT scan- breath-hold level, accurate targeting of smaller nodules ner, all measurements were performed on a regular becomes impossible. table, with a setup simulating that of a CT table with CT Currently available respiratory tracking systems suit- gantry (Fig. 2). able for image-guided intervention consist of respiratory belts that are cumbersome to install, only have a weak Software correlation with nodule position and do not adjust for a The Kinect provides a depth map and a colour image of change in breathing pattern [2]. Several groups have in- the scene. Using Matlab 2014a’s Image Acquisition Tool- vestigated the use of a depth camera to monitor patient kit (the Mathworks, Natick, MA, USA) and an in-house respiratory motion for four-dimensional radiotherapy written script, these feeds were processed. The operator planning [3–6]. Depth cameras measure the distance to interacted with this script using a graphical user inter- a surface for each pixel and can thus be used to deter- face. First, the colour image was used to interactively mine changes in skin surface, caused by respiratory select a polygonal region of interest (ROI) outlining the motion, in real time [7]. abdomen and thorax, including as much as possible In this study, we implemented and tested a similar from the visible skin surface. This step takes approxi- setup, in combination with real-time visual biofeedback mately 5 s. Figure 3 shows a screenshot of the ROI to the patient. The aim was to present and evaluate a selection process. Erroneous data pixels (identified as new respiratory level biofeedback system that aids distance = 0) were excluded and the mean distance to patients to return to a consistent level of breath-hold the entire ROI was calculated. The mean distance and with potential application in image-guided interventions. corresponding time point were saved for every frame, with a frame rate of approximately 15 Hz. After the Methods selection of the ROI, the preferred level of expiration was set using the graphical user interface. When this The setup was selected, the volunteers received visual biofeedback of The Kinect for Windows V1 (Microsoft, Redmond, VA, the level of expiration: on a screen next to the volunteer, a USA) was used to measure respiratory surface move- red circle moved up and down with respiration and a ment. It can be positioned onto the CT table using a green circle corresponded to the respiratory level, setup with an adapted tripod as shown in Fig. 1, where it previously set (Fig. 4). remains stationary relative to the patient. The Kinect Volunteer experiments The study has been reviewed and the need for written informed consent was waived by the Medical Ethics Review Board of University Medical Center Groningen (number 2017/226). Initial validation of the accuracy of Fig. 1 CT table setup. Setup of the Kinect camera on a CT table, Fig. 2 Experimental setup. Setup of how the Kinect was used in this positioned with its field of view into the gantry. The tablet can study. Volunteers could see the visual biofeedback on the additional provide visual biofeedback to patient and operator screen Heerink et al. European Radiology Experimental (2018) 2:22 Page 3 of 9 Fig. 3 Screenshot of ROI selection process. The colour image is captured by the Kinect and the blue lines represent the border of the selected ROI Fig. 4 Screenshot of biofeedback provided to the volunteers. The red circle moves up and down with respiration and the green circle the depth measurements as a measure of respiratory corresponds to the respiratory level, previously set level was performed on a single healthy volunteer. He was positioned on a vacuum mattress (BodyFIX Blue- BAG, Elekta, Crawley, UK) to limit movement and of the mattress while making sure the volunteers’ sides maximise the respiratory surface motion on the skin were fully supported by the mattress, too. surface and asked to perform several respiratory ma- Volunteers were randomly positioned in the supine or noeuvres, in the supine and prone positions. During prone position first. Next, a conversion factor from mm these manoeuvres, the mean depth measurements were to mL was determined before receiving biofeedback. recorded and a Jaeger Masterscope with spirometry This was performed by measuring the change in lung software (SentrySuite V2.13, CareFusion, San Diego, volume simultaneously with the change in mean CA, USA) was used to measure the respiratory level. distance to the ROI. The minimal and maximal values of The spirometer was calibrated using a 3 L calibration both signals of four corresponding tidal volumes around syringe. Local ambient temperature, humidity and air functional residual capacity (FRC) level were extrapo- pressure were updated in the system. The respiratory lated from the graphs and averaged. The mean tidal vol- manoeuvres consisted of a complete inspiration, result- ume (in mL) was subsequently divided by the mean tidal ing in a peak in both signals to temporally align them, movement (in mm) to determine a conversion factor C and several tidal volumes. The volume and inverted (mL/mm) for every volunteer for both positions. mean distance measurements for the supine and prone Subsequently, without spirometry, the volunteers were positions were superimposed plotted in graphs and given breath-hold instructions: ‘Breathe out…, breathe assessed visually. in…, breathe out and hold your breath’. This way, the To evaluate whether the visual biofeedback can facili- volunteers held their breath at FRC, approximately. In tate the return to a predictable and consistent level of fact, breath-hold at a lower lung volume results in less breath-hold, the system was tested with eight healthy organ motion compared to inspiration, because of a volunteers (four men, four women). All volunteers had decreased gas exchange in the alveoli [8–10]. Addition- their height and weight measured and their body mass ally, the British Thoracic Society advises breath-hold at index was calculated. After the volunteers were posi- FRC level (gentle expiration) for biopsy of nodules in the tioned correctly on the table, the air was vacuumed out lung base [11]. Heerink et al. European Radiology Experimental (2018) 2:22 Page 4 of 9 Then, the volunteers were asked ten times to return to Results the same level of breath-hold for approximately 5 s, with The median age of the volunteers was 29 years (IQR 25– 30s intervals. This was performed twice, in a random 33 years), with a median body mass index of 21.5 kg/m order: once while the volunteers received the same (IQR 20.9–23.3 kg/m ). Figures 5 and 6 show the graphs breathing instruction but could not see the visual bio- of the initial validation measurements of a single volunteer feedback; and once while they could see the screen with in the supine and prone positions, respectively. The biofeedback. They were asked to slowly breathe out, to measurements performed with the Kinect system and with prevent any overshoot, until the red ball was inside the the spirometer are superimposed on each other. green circle and then hold their breath. Additionally, All parameters were non-normally distributed (p < 0.001). they were instructed not to correct their breath-hold in The conversion factors did not differ between volunteer case they did overshoot the target level. positions (C = 187.8 mL/mm, C = 164.7 mL/mm; supine prone p = 0.767) or between men and women (C = 174.1 mL/ males Variables and statistical analysis mm, C = 158.9 mL/mm; p = 0.452). females Median conversion factors were determined for volunteer Figures 7 and 8 show graphs of the respiratory level of position (C and C ) and sex (C and C ). a volunteer (F4) in the prone position, without and with supine prone males females The periods of breath-hold were selected and their mean biofeedback, respectively. Figures 9 and 10 show box values determined using a graphical data selection tool in plots of the all the individual volunteers, without and Matlab. The absolute error (E) between the initial mean with biofeedback, grouped by volunteer, for the supine level of breath-hold and the consequent attempts to re- and prone positions, respectively. turn to said level were saved. These values were converted For all volunteers, the absolute error from initial from distance to volume using the corresponding conver- breath-hold at FRC level reduced from 147.6 mL (IQR sion factor. 76.8–276.8 mL) without feedback to 27.7 mL (IQR Normality of data distribution was evaluated with the 10.9–51.7 mL) with feedback (p < 0.001). Table 1 shows Shapiro-Wilk test. Due to non-normal distribution, the absolute error for both sexes, in both positions, median and interquartile range (IQR) of the absolute stratified by feedback. The error without feedback was error were determined for the measurements with and higher in the prone position (E = 156 mL vs E = prone supine without feedback overall, by sex, by position (supine and 125 mL, p = 0.012). With feedback, there was no signifi- prone) and by volunteer. Paired variables (with feedback cant difference between the two volunteer positions, vs without feedback) were compared using Wilcoxon though the error trended to be lower in the prone signed rank tests, paired by the volunteer’s breath-hold position (E =22 mL vs E =32 mL, p = 0.086). prone supine attempt and unpaired variables (men vs women, supine vs prone) were compared using Mann-Whitney U tests. Discussion Box plots were made for the results in the supine and The aim of this study was to present and evaluate a new prone positions, grouped per volunteer, comparing with respiratory level biofeedback system that aids patients to and without feedback. return to a consistent level of breath-hold with potential All measurements were performed in Matlab and application to image-guided interventions. We demon- results were subsequently imported in SPSS 23.0 (IBM, strated that the system described in this paper enables New York, NY, USA) to perform statistical analysis. healthy volunteers to return to 28 mL of their initial Level of significance was set at p < 0.050. breath-hold, which is a significant reduction from the 3.5 -1150 2.5 -1155 1.5 -1160 VT 0.5 FRC -1165 25 30 35 40 45 Time (s) Fig. 5 Supine position. Superimposed graphs of mean depth, measured by the Kinect (blue, left axis) and volume, measured with spirometry (red, right axis) of a volunteer in the supine position. Periods of breath-hold are indicated with grey boxes. FRC functional residual capacity, VT tidal volume Depth (mm) Volume (L) Heerink et al. European Radiology Experimental (2018) 2:22 Page 5 of 9 1124 3 2.5 1126 2.5 1128 2 2 1130 1.5 1.5 1 1 1134 VT 0.5 0.5 1136 FRC 0 0 15 20 20 25 30 35 40 -0.5 -0.5 Time (s) Fig. 6 Prone position. Superimposed graphs of mean depth, measured by the Kinect (blue, left axis) and volume, measured with spirometry (red, right axis) of a volunteer in prone position. Periods of breath-hold are indicated with grey boxes. FRC functional residual capacity, VT tidal volume 147 mL that they managed without the biofeedback When targeting a lesion in image-guided interven- system. tions, it is not really a reliable, consistent lung volume Without feedback, the volunteers had a larger absolute that is important. For a radiologist, it is about the target error in the prone position compared to the supine pos- being in the same position to when the image was ition (E = 156 mL vs E = 125 mL, p = 0.012). acquired. However, to analyse whether the system pre- prone supine This illustrates that it is harder for patients to get to the sented here results in a reproducible target position same level of breath-hold while lying on their stomach. would require using a CT scanner, resulting in a radi- With feedback, the volunteers no longer had increased ation dose in healthy volunteers. Therefore, this was not difficulty with the prone position compared to the an option for this study. Using CT, Chen et al. [1] inves- supine position. In fact, they seemed to perform better tigated the motion of lung nodules from full inspiration in the prone position (E =22 mL vs E =32 mL, to end-expiration during tidal volume breathing (i.e. prone supine p = 0.086). We speculate this is because the back pro- inspiratory capacity). The average motion of all 85 vides a more stable platform to measure a mean dis- included nodules was 17.6 mm; in the left and right tance to because there is less soft tissue such as fat lower lobes, this was 23.8 mm and 25.3 mm, respect- and breast tissue to impair the measurements. In our ively. The average inspiratory capacity of men and department, approximately half of the CT-guided lung women was 3.5 L and 2.4 L, so considering a linear biopsies are performed in the prone position, so for relation, a lung volume change of 100 mL would result these procedures visual biofeedback will be of in- in a nodule motion of 1.1 mm and 0.7 mm in the lower creased importance. lobes, for men and women, respectively [12]. Translating 0 50 100 150 200 250 300 350 400 Time (s) Fig. 7 Respiratory level without feedback. Respiratory level of a volunteer (F4) in the prone position, without visual feedback. Red line represents the initial level of breath-hold. The volunteer was given breathing instruction every 30 s and asked to return to the same respiratory level and maintain breath-hold for a couple of s, indicated by the short horizontal periods Depth (mm) Mean distance (mm) Volume (L) Heerink et al. European Radiology Experimental (2018) 2:22 Page 6 of 9 0 50 100 150 200 250 300 350 Time (s) Fig. 8 Respiratory level with feedback. Respiratory level of a volunteer (F4) in the prone position, with visual feedback. Red line is the initial level of breath-hold. Volunteer was asked to return to the same respiratory level every 30 s with the aid of visual biofeedback and maintain breath-hold for a couple of s, indicated by the short horizontal periods these numbers to the results of this study, one can con- respiratory motion from the study by Chen et al. [1] (up clude that even in the lower lung lobes, the biofeedback to 60 mm), this would result in a reproducibility of the system can potentially enable men and women to have a nodule position of within 1 mm. predictable consistent nodule position of well below Price et al. [6] have recently performed a clinical trial 0.5 mm. Even when considering nodules with extreme to assess the feasibility of using an in-house developed Position: Prone Without feedback With feedback F1 F2 F3 F4 M1 M2 M3 M4 Volunteer Fig. 9 Breath-hold error in prone position. Box plots of the error from initial breath-hold, without and with biofeedback, grouped by volunteer, in the prone position Mean distance (mm) Absolute error from inital breath-hold (mL) Heerink et al. European Radiology Experimental (2018) 2:22 Page 7 of 9 Position: Supine Without feedback With feedback F1 F2 F3 F4 M1 M2 M3 M4 Volunteer Fig. 10 Breath-hold in supine position. Box plots of the error from initial breath-hold, without and with biofeedback, grouped by volunteer, in the supine position optical surface tracking device to facilitate consistent study, the skin movement of the volunteers improved breath-hold during radiation therapy. They found that from 0.79 to 0.15 mm, when providing the feedback. patients were able to tolerate the feedback well and that Though these measurements cannot be directly com- they had a moderately improved reproducibility of skin pared, because they rely on technical factors as camera surface. They used traffic light colours to provide visual angle, our system should have a relatively higher rate of feedback to the patients and were able to reduce the improvement. mean amplitude of skin movement from 2.0 mm to The Kinect has the added benefit of being a generally 1.7 mm. In a previous healthy volunteer study [13], they available, low-cost system. Several groups have analysed achieved an improvement from 1.4 to 0.6 mm. In our the feasibility of using the Kinect camera to monitor re- spiratory motion for respiratory gated or four-dimensional CT-based continuous radiotherapy [3–5]. Though the Table 1 Absolute error from initial breath-hold at expiration results seem promising, to our knowledge, no clinical Median Q1–Q3 p value studies utilising the Kinect have been published yet. The Skin surface error without feedback (mm) 0.79 0.47–1.56 < 0.001 Kinect-based respiratory motion monitoring systems are Skin surface error feedback (mm) 0.15 0.06–0.30 mostly compared with the RPM Gating System (Varian E without feedback (mL) 147 76.8–276.8 < 0.001 overall Medical System, Palo Alto, CA, USA), a clinically available E with feedback (mL) 27.7 10.9–51.7 overall respiratory motion tracking system that utilised the move- ment of a marker box placed on the patient’s chest to gate E with feedback (mL) 35.5 14.3–52.3 0.081 men radiation therapy. This system is not suitable for interven- E with feedback (mL) 23.7 8.2–50.4 women tional procedures because the box has to be placed on E without feedback (mL) 124.6 55.7–259.7 0.012 supine disinfected skin and can easily be knocked out of place. As E without feedback (mL) 156.3 90.9–334.7 prone it only tracks the movement of a single marker, it would E with feedback (mL) 31.7 12.8–59.6 0.086 supine not be able to detect a change in breathing pattern either. E with feedback (mL) 22.3 7.7–47.0 prone During interventional procedures, patients are more likely E absolute error from initial breath-hold at functional residual capacity level to alter from thoracic to abdominal breathing, or vice Absolute error from inital breath-hold (mL) Heerink et al. European Radiology Experimental (2018) 2:22 Page 8 of 9 versa, rendering the tracking inaccurate. The value of skin glottis. Moreover, the spirometer data showed a signifi- surface motion tracking in combination with a tightly po- cant drift in the long-term measurements, even after sitioned vacuum mattress is that all respiratory movement rigorous (re)calibration of the spirometer, rendering can be visualised and thus be used as patient feedback. these long-term spirometry data unusable. Of note, the A change in breathing pattern is also a problem when spirometry measurements are not required for the sys- using abdominal/chest belts. These belts measure the tem to function in clinical practice. These were per- circumference of the patient’s chest or abdomen to formed only for validation of the system. provide patient feedback. Schoth et al. [14, 15] reported There are some limitations to this study. With a mean reduced intervention time and radiation exposure using age of 29 years and a mean body mass index of 22.2, the the IBC system (Mayo Clinic Medical Devices, USA) for volunteers were all young healthy adults, compared to the CT-guided lung biopsy while Carlson et al. [14, 15] potential target group. The breathing instructions resulted reported a reduction in targeting attempts using this belt in a breath-hold at FRC level. Obese patients generally in CT fluoroscopy-guided lung biopsy. However, in our breathe at a lower FRC [18] and patients suffering from experience these belts are cumbersome to setup and chronic obstructive pulmonary disease breathe at a higher unreliable. In a review of another bellows belt system FRC than healthy volunteers [19]. Although the relation (Philips Medical Systems, Eindhoven, The Netherlands), between chest wall motion and diaphragmatic excursion is Locklin et al. [2] demonstrated only a weak correlation approximately linear in healthy adults, this might not be between chest circumference and nodule position. true for patients [20]. Additionally, Harte et al. [21]have Another option that has been considered is the use of shown that patients with cystic fibrosis have a lower cor- spirometry to monitor lung volume. Tomiyama et al. [16] relation between chest wall movement and lung volume used a respiratory monitor to trigger an electric light bulb changes. Although differences in tracking accuracy are to as an indication of a similar level of breath-hold, resulting be expected between the volunteers and patients, the pro- in a high diagnostic accuracy (96%) in CT-guided biopsy cedure with biofeedback is feasible with low error rates of small (< 15 mm) lung nodules. The active breathing co- and easy to instruct to volunteers. ordinator system (Elekta Instrument AB, Stockholm, Patients might arguably have more difficulty in inter- Sweden) is a clinically available system used to actively preting the biofeedback and therefore in returning to monitor lung volume and suspend the patient’s breathing. their initial level of breath-hold every time. This does A valve closes and holds respiration at a certain level, to not have to lead to targeting errors per se, because the facilitate consisted tumour position for gated radiotherapy. operator will also see the biofeedback. He/she can there- From a practical standpoint, spirometry seems less suited fore keep on instructing the patient until the patient for interventional procedures, because it prevents com- manages to hold his breath is at the level the CT scan munication from patient to physician. If any breath were was acquired, before proceeding with needle manipula- to escape from the mouthpiece, the procedure would no tion. It should also be considered that patients who are longer be reliable. difficult to instruct will have more difficulty to return to As an alternative to CT-guided lung biopsy, CT fluor- a consistent level of breath-hold with only breathing oscopy can be used to target lung lesions. The advantage instructions, so these patients might benefit even more of CT fluoroscopy is the real-time feedback it provides from the feedback system. so the radiologist can verify the lesions’ position and im- Often, in interventional procedures parts of the thorax mediately place the biopsy needle. It has been reported and abdomen have to be covered with sterile drapes. with similar diagnostic accuracy and lower complication Movement of these drapes would result in an error in the rates, compared to CT-guided lung biopsy [17]. How- depth measurements. This can be prevented by either ever, both patient and radiologist experience a higher ex- using a drape with a large hole, and disinfecting a larger posure to radiation, which is a significant disadvantage. surface of the skin, or by using surgical incise drape with Additionally, depending on the CT scanner’s gantry tilt, adhesive backing. This facilitates a larger ROI to be the biopsy needle path is limited to in-plane approaches. selected, without having to include loose fitting drapes. During the biofeedback evaluation, the volunteers In conclusion, we presented a method to provide were not breathing into a spirometer. Instead, the patients with visual biofeedback of their respiratory level, spirometer was used during validation measurements to enable them to return to a consistent level of beforehand, to determine a conversion factor from chest breath-hold during image-guided interventions. The depth height to lung volume. When we attempted to use the measurements have proven to be an accurate measure of spirometer during the feedback evaluation, some volun- lung volume and the visual biofeedback enabled healthy teers demonstrated difficulty in maintaining breath-hold volunteers to return to 28 mL of their initial breath-hold because the mount piece prevented them from closing at expiratory level, corresponding to an estimated target their mouth, as if they were not able to close their position reproducibility of < 0.5 mm. If implemented in an Heerink et al. European Radiology Experimental (2018) 2:22 Page 9 of 9 image-guided intervention suite, it has the potential to 8. Holland AE, Goldfarb JW, Edelman RR (1998) Diaphragmatic and cardiac motion during suspended breathing: preliminary experience and prevent targeting errors caused by respiratory motion and implications for breath-hold MR imaging. Radiology 209:483–489 thereby to increase targeting accuracy. 9. Lanphier EH, Rahn H (1963) Alveolar gas exchange during breath holding with air. J Appl Physiol 18:478–482 Abbreviations 10. Lens E, Gurney-Champion OJ, Tekelenburg DR et al (2016) Abdominal organ C: Conversion factor; CT: Computed tomography; D: Difference; E: Absolute motion during inhalation and exhalation breath-holds: pancreatic motion at error; FRC: Functional residual capacity; IQR: Interquartile range; ROI: Region different lung volumes compared. Radiother On of interest 11. Manhire A, Charig M, Clelland C et al (2003) Guidelines for radiologically guided lung biopsy. Thorax 58:920–936 Availability of data and materials 12. Barret KE, Ganong WF (2012) Ganong’s review of medical physiology, 25th Data are available online in the Open Science Framework data repository edn. McGraw-Hill Medical, New York (https://osf.io/5ysfv/). 13. Parkhurst JM, Price GJ, Sharrock PJ, Jackson ASN, Stratford J, Moore CJ (2013) Self-Management of Patient Body Position, pose, and motion using wide-field, Funding real-time optical measurement feedback: results of a volunteer study. Int J This study was funded by Samenwerkingsverband Noord Nederland. Radiat Oncol Biol Phys 87:904–910 14. Schoth F, Plumhans C, Kraemer N et al (2010) Evaluation of an interactive Acknowledgments breath-hold control system in CT-guided lung biopsy. Rofo 182:507–511 Not applicable 15. Carlson SK, Felmlee JP, Bender CE et al (2005) CT fluoroscopy–guided biopsy of the lung or upper abdomen with a breath-hold monitoring and Authors’ contributions feedback system: a prospective randomized controlled clinical trial. WJH designed and conducted the study. WJH performed data analysis. WJH, Radiology 237:701–708 RV, PvO and MO performed data interpretation. WJH wrote the paper and all 16. Tomiyama N, Mihara N, Maeda M et al (2000) CT-guided needle biopsy of authors provided critical revisions and gave final approval. small pulmonary nodules: value of respiratory gating. Radiology 217:907–910 17. Kim GR, Hur J, Lee SM et al (2011) CT fluoroscopy-guided lung biopsy versus Ethics approval and consent to participate conventional CT-guided lung biopsy: a prospective controlled study to assess The study was approved by the local ethics committee and written informed radiation doses and diagnostic performance. Eur Radiol 21:232–239 consent was waived. 18. Parameswaran K, Todd DC, Soth M (2006) Altered respiratory physiology in obesity. Can Respir J 13:203–210 Consent for publication 19. Barnes PJ, Drazen JM, Rennard SI, Thomson NC (2009) Asthma and COPD Consent for publication has been provided by the volunteer pictured in basic mechanisms and clinical management. Elsevier Ltd, San Diego Fig. 2 and by the volunteer whose feedback is presented in Figs. 5 and 6. 20. Wang H-K, Lu T-W, Liing R-J, Shih TT-F, Chen S-C, Lin K-H (2009) Relationship between Chest Wall motion and diaphragmatic excursion in Competing interests healthy adults in supine position. J Formos Med Assoc 108:577–586 The authors declare that they have no competing interests. 21. Harte JM, Golby CK, Acosta J et al (2016) Chest wall motion analysis in healthy volunteers and adults with cystic fibrosis using a novel Kinect-based motion tracking system. Med Biol Eng Comput 54:1631–1640 Publisher’sNote Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Author details Center for Medical Imaging – North East Netherlands, University of Groningen, Groningen, The Netherlands. Department of Radiology, University of Groningen, University Medical Center Groningen, Groningen, The Netherlands. Department of Pulmonary Diseases, University of Groningen, University Medical Center Groningen, Groningen, The Netherlands. Received: 23 February 2018 Accepted: 29 May 2018 References 1. Chen A, Pastis N, Furukawa B, Silvestri GA (2015) The effect of respiratory motion on pulmonary nodule location during electromagnetic navigation bronchoscopy. Chest 147:1275–1281 2. Locklin JK, Yanof J, Luk A, Varro Z, Patriciu A, Wood BJ (2007) Respiratory biofeedback during CT-guided procedures. J Vasc Interv Radiol 18:749–755 3. Tahavori F, Alnowami M, Wells K (2014) Marker-less respiratory motion modeling using the Microsoft Kinect for windows. SPIE Medical Imaging 1–10 4. Lim SH, Golkar E, Rahni AAA (2014) Respiratory motion tracking using the Kinect camera. IEEE Conference on Biomedical Engineering and Sciences (IECBES). https://doi.org/10.1109/IECBES.2014.7047619 5. Ortmüller J, Gauer T, Wilms M, Handels H, Werner R (2015) Respiratory surface motion measurement by Microsoft Kinect. Current Directions in Biomedical Engineering 1:270–273 6. Price GJ, Faivre-Finn C, Stratford J et al (2017) Results from a clinical trial evaluating the efficacy of real-time body surface visual feedback in reducing patient motion during lung cancer radiotherapy. Acta Oncol 57:211–218 7. Liu HH, Koch N, Starkschall G et al (2004) Evaluation of internal lung motion for respiratory-gated radiotherapy using MRI: part II—margin reduction of internal target volume. Int J Radiat Oncol Biol Phys. 60:1473–1483 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png European Radiology Experimental Springer Journals

Respiratory level tracking with visual biofeedback for consistent breath-hold level with potential application in image-guided interventions

Download PDF
9 pages

Loading next page...
 
/lp/springer-journals/respiratory-level-tracking-with-visual-biofeedback-for-consistent-A0s74Obykz

References (40)

Publisher
Springer Journals
Copyright
Copyright © 2018 by The Author(s)
Subject
Medicine & Public Health; Imaging / Radiology; Diagnostic Radiology; Interventional Radiology; Neuroradiology; Ultrasound; Internal Medicine
eISSN
2509-9280
DOI
10.1186/s41747-018-0052-7
Publisher site
See Article on Publisher Site

Abstract

Background: To present and evaluate a new respiratory level biofeedback system that aids the patient to return to a consistent breath-hold level with potential application in image-guided interventions. Methods: The study was approved by the local ethics committee and written informed consent was waived. Respiratory motion was recorded in eight healthy volunteers in the supine and prone positions, using a depth camera that measures the mean distance to thorax, abdomen and back. Volunteers were provided with real-time visual biofeedback on a screen, as a ball moving up and down with respiratory motion. For validation purposes, a conversion factor from mean distance (in mm) to relative lung volume (in mL) was determined using spirometry. Subsequently, without spirometry, volunteers were given breathing instructions and were asked to return to their initial breath-hold level at expiration ten times, in both positions, with and without visual biofeedback. For both positions, the median and interquartile range (IQR) of the absolute error in lung volume from initial breath-hold were determined with and without biofeedback and compared using Wilcoxon signed rank tests. Results: Without visual biofeedback, the median difference from initial breath-hold was 124.6 mL (IQR 55.7–259.7 mL) for the supine position and 156.3 mL (IQR 90.9–334.7 mL) for the prone position. With the biofeedback, the difference was significantly decreased to 32.7 mL (IQR 12.8–59.6 mL) (p < 0.001) and 22.3 mL (IQR 7.7–47.0 mL) (p <0.001), respectively. Conclusions: The use of a depth camera to provide visual biofeedback increased the reproducibility of breath-hold expiration level in healthy volunteers, with a potential to eliminate targeting errors caused by respiratory movement during lung image-guided procedures. Keywords: Breath holding, Image-guided biopsy, Lung, Reproducibilty of results, Respiration Key points  This method can have potential application in lung image-guided interventional procedures, reducing A depth camera can be used to accurately monitor targeting errors caused by respiration the level of respiration Visual feedback enables volunteers to hold their Background breath at a consistent level Tumour movement caused by patient respiration can be a serious problem during image-guided interventional procedures. Lung nodules near the diaphragm, for * Correspondence: w.j.heerink@umcg.nl Center for Medical Imaging – North East Netherlands, University of example, typically move > 2 cm with inspiratory capacity Groningen, Groningen, The Netherlands [1]. When tissue diagnosis of a suspicious lung nodule Department of Radiology, University of Groningen, University Medical by computed tomography (CT)-guided biopsy is Center Groningen, Groningen, The Netherlands Full list of author information is available at the end of the article required, a predictable nodule position is important. © 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. Heerink et al. European Radiology Experimental (2018) 2:22 Page 2 of 9 Often, giving a patient specific breathing instructions is was angled forward towards the patient, so the patient not sufficient to facilitate a consistent breath-hold level. remained in its field of view, while in the CT gantry. For When patients have difficulty returning to their initial convenience and lack of continuous access to a CT scan- breath-hold level, accurate targeting of smaller nodules ner, all measurements were performed on a regular becomes impossible. table, with a setup simulating that of a CT table with CT Currently available respiratory tracking systems suit- gantry (Fig. 2). able for image-guided intervention consist of respiratory belts that are cumbersome to install, only have a weak Software correlation with nodule position and do not adjust for a The Kinect provides a depth map and a colour image of change in breathing pattern [2]. Several groups have in- the scene. Using Matlab 2014a’s Image Acquisition Tool- vestigated the use of a depth camera to monitor patient kit (the Mathworks, Natick, MA, USA) and an in-house respiratory motion for four-dimensional radiotherapy written script, these feeds were processed. The operator planning [3–6]. Depth cameras measure the distance to interacted with this script using a graphical user inter- a surface for each pixel and can thus be used to deter- face. First, the colour image was used to interactively mine changes in skin surface, caused by respiratory select a polygonal region of interest (ROI) outlining the motion, in real time [7]. abdomen and thorax, including as much as possible In this study, we implemented and tested a similar from the visible skin surface. This step takes approxi- setup, in combination with real-time visual biofeedback mately 5 s. Figure 3 shows a screenshot of the ROI to the patient. The aim was to present and evaluate a selection process. Erroneous data pixels (identified as new respiratory level biofeedback system that aids distance = 0) were excluded and the mean distance to patients to return to a consistent level of breath-hold the entire ROI was calculated. The mean distance and with potential application in image-guided interventions. corresponding time point were saved for every frame, with a frame rate of approximately 15 Hz. After the Methods selection of the ROI, the preferred level of expiration was set using the graphical user interface. When this The setup was selected, the volunteers received visual biofeedback of The Kinect for Windows V1 (Microsoft, Redmond, VA, the level of expiration: on a screen next to the volunteer, a USA) was used to measure respiratory surface move- red circle moved up and down with respiration and a ment. It can be positioned onto the CT table using a green circle corresponded to the respiratory level, setup with an adapted tripod as shown in Fig. 1, where it previously set (Fig. 4). remains stationary relative to the patient. The Kinect Volunteer experiments The study has been reviewed and the need for written informed consent was waived by the Medical Ethics Review Board of University Medical Center Groningen (number 2017/226). Initial validation of the accuracy of Fig. 1 CT table setup. Setup of the Kinect camera on a CT table, Fig. 2 Experimental setup. Setup of how the Kinect was used in this positioned with its field of view into the gantry. The tablet can study. Volunteers could see the visual biofeedback on the additional provide visual biofeedback to patient and operator screen Heerink et al. European Radiology Experimental (2018) 2:22 Page 3 of 9 Fig. 3 Screenshot of ROI selection process. The colour image is captured by the Kinect and the blue lines represent the border of the selected ROI Fig. 4 Screenshot of biofeedback provided to the volunteers. The red circle moves up and down with respiration and the green circle the depth measurements as a measure of respiratory corresponds to the respiratory level, previously set level was performed on a single healthy volunteer. He was positioned on a vacuum mattress (BodyFIX Blue- BAG, Elekta, Crawley, UK) to limit movement and of the mattress while making sure the volunteers’ sides maximise the respiratory surface motion on the skin were fully supported by the mattress, too. surface and asked to perform several respiratory ma- Volunteers were randomly positioned in the supine or noeuvres, in the supine and prone positions. During prone position first. Next, a conversion factor from mm these manoeuvres, the mean depth measurements were to mL was determined before receiving biofeedback. recorded and a Jaeger Masterscope with spirometry This was performed by measuring the change in lung software (SentrySuite V2.13, CareFusion, San Diego, volume simultaneously with the change in mean CA, USA) was used to measure the respiratory level. distance to the ROI. The minimal and maximal values of The spirometer was calibrated using a 3 L calibration both signals of four corresponding tidal volumes around syringe. Local ambient temperature, humidity and air functional residual capacity (FRC) level were extrapo- pressure were updated in the system. The respiratory lated from the graphs and averaged. The mean tidal vol- manoeuvres consisted of a complete inspiration, result- ume (in mL) was subsequently divided by the mean tidal ing in a peak in both signals to temporally align them, movement (in mm) to determine a conversion factor C and several tidal volumes. The volume and inverted (mL/mm) for every volunteer for both positions. mean distance measurements for the supine and prone Subsequently, without spirometry, the volunteers were positions were superimposed plotted in graphs and given breath-hold instructions: ‘Breathe out…, breathe assessed visually. in…, breathe out and hold your breath’. This way, the To evaluate whether the visual biofeedback can facili- volunteers held their breath at FRC, approximately. In tate the return to a predictable and consistent level of fact, breath-hold at a lower lung volume results in less breath-hold, the system was tested with eight healthy organ motion compared to inspiration, because of a volunteers (four men, four women). All volunteers had decreased gas exchange in the alveoli [8–10]. Addition- their height and weight measured and their body mass ally, the British Thoracic Society advises breath-hold at index was calculated. After the volunteers were posi- FRC level (gentle expiration) for biopsy of nodules in the tioned correctly on the table, the air was vacuumed out lung base [11]. Heerink et al. European Radiology Experimental (2018) 2:22 Page 4 of 9 Then, the volunteers were asked ten times to return to Results the same level of breath-hold for approximately 5 s, with The median age of the volunteers was 29 years (IQR 25– 30s intervals. This was performed twice, in a random 33 years), with a median body mass index of 21.5 kg/m order: once while the volunteers received the same (IQR 20.9–23.3 kg/m ). Figures 5 and 6 show the graphs breathing instruction but could not see the visual bio- of the initial validation measurements of a single volunteer feedback; and once while they could see the screen with in the supine and prone positions, respectively. The biofeedback. They were asked to slowly breathe out, to measurements performed with the Kinect system and with prevent any overshoot, until the red ball was inside the the spirometer are superimposed on each other. green circle and then hold their breath. Additionally, All parameters were non-normally distributed (p < 0.001). they were instructed not to correct their breath-hold in The conversion factors did not differ between volunteer case they did overshoot the target level. positions (C = 187.8 mL/mm, C = 164.7 mL/mm; supine prone p = 0.767) or between men and women (C = 174.1 mL/ males Variables and statistical analysis mm, C = 158.9 mL/mm; p = 0.452). females Median conversion factors were determined for volunteer Figures 7 and 8 show graphs of the respiratory level of position (C and C ) and sex (C and C ). a volunteer (F4) in the prone position, without and with supine prone males females The periods of breath-hold were selected and their mean biofeedback, respectively. Figures 9 and 10 show box values determined using a graphical data selection tool in plots of the all the individual volunteers, without and Matlab. The absolute error (E) between the initial mean with biofeedback, grouped by volunteer, for the supine level of breath-hold and the consequent attempts to re- and prone positions, respectively. turn to said level were saved. These values were converted For all volunteers, the absolute error from initial from distance to volume using the corresponding conver- breath-hold at FRC level reduced from 147.6 mL (IQR sion factor. 76.8–276.8 mL) without feedback to 27.7 mL (IQR Normality of data distribution was evaluated with the 10.9–51.7 mL) with feedback (p < 0.001). Table 1 shows Shapiro-Wilk test. Due to non-normal distribution, the absolute error for both sexes, in both positions, median and interquartile range (IQR) of the absolute stratified by feedback. The error without feedback was error were determined for the measurements with and higher in the prone position (E = 156 mL vs E = prone supine without feedback overall, by sex, by position (supine and 125 mL, p = 0.012). With feedback, there was no signifi- prone) and by volunteer. Paired variables (with feedback cant difference between the two volunteer positions, vs without feedback) were compared using Wilcoxon though the error trended to be lower in the prone signed rank tests, paired by the volunteer’s breath-hold position (E =22 mL vs E =32 mL, p = 0.086). prone supine attempt and unpaired variables (men vs women, supine vs prone) were compared using Mann-Whitney U tests. Discussion Box plots were made for the results in the supine and The aim of this study was to present and evaluate a new prone positions, grouped per volunteer, comparing with respiratory level biofeedback system that aids patients to and without feedback. return to a consistent level of breath-hold with potential All measurements were performed in Matlab and application to image-guided interventions. We demon- results were subsequently imported in SPSS 23.0 (IBM, strated that the system described in this paper enables New York, NY, USA) to perform statistical analysis. healthy volunteers to return to 28 mL of their initial Level of significance was set at p < 0.050. breath-hold, which is a significant reduction from the 3.5 -1150 2.5 -1155 1.5 -1160 VT 0.5 FRC -1165 25 30 35 40 45 Time (s) Fig. 5 Supine position. Superimposed graphs of mean depth, measured by the Kinect (blue, left axis) and volume, measured with spirometry (red, right axis) of a volunteer in the supine position. Periods of breath-hold are indicated with grey boxes. FRC functional residual capacity, VT tidal volume Depth (mm) Volume (L) Heerink et al. European Radiology Experimental (2018) 2:22 Page 5 of 9 1124 3 2.5 1126 2.5 1128 2 2 1130 1.5 1.5 1 1 1134 VT 0.5 0.5 1136 FRC 0 0 15 20 20 25 30 35 40 -0.5 -0.5 Time (s) Fig. 6 Prone position. Superimposed graphs of mean depth, measured by the Kinect (blue, left axis) and volume, measured with spirometry (red, right axis) of a volunteer in prone position. Periods of breath-hold are indicated with grey boxes. FRC functional residual capacity, VT tidal volume 147 mL that they managed without the biofeedback When targeting a lesion in image-guided interven- system. tions, it is not really a reliable, consistent lung volume Without feedback, the volunteers had a larger absolute that is important. For a radiologist, it is about the target error in the prone position compared to the supine pos- being in the same position to when the image was ition (E = 156 mL vs E = 125 mL, p = 0.012). acquired. However, to analyse whether the system pre- prone supine This illustrates that it is harder for patients to get to the sented here results in a reproducible target position same level of breath-hold while lying on their stomach. would require using a CT scanner, resulting in a radi- With feedback, the volunteers no longer had increased ation dose in healthy volunteers. Therefore, this was not difficulty with the prone position compared to the an option for this study. Using CT, Chen et al. [1] inves- supine position. In fact, they seemed to perform better tigated the motion of lung nodules from full inspiration in the prone position (E =22 mL vs E =32 mL, to end-expiration during tidal volume breathing (i.e. prone supine p = 0.086). We speculate this is because the back pro- inspiratory capacity). The average motion of all 85 vides a more stable platform to measure a mean dis- included nodules was 17.6 mm; in the left and right tance to because there is less soft tissue such as fat lower lobes, this was 23.8 mm and 25.3 mm, respect- and breast tissue to impair the measurements. In our ively. The average inspiratory capacity of men and department, approximately half of the CT-guided lung women was 3.5 L and 2.4 L, so considering a linear biopsies are performed in the prone position, so for relation, a lung volume change of 100 mL would result these procedures visual biofeedback will be of in- in a nodule motion of 1.1 mm and 0.7 mm in the lower creased importance. lobes, for men and women, respectively [12]. Translating 0 50 100 150 200 250 300 350 400 Time (s) Fig. 7 Respiratory level without feedback. Respiratory level of a volunteer (F4) in the prone position, without visual feedback. Red line represents the initial level of breath-hold. The volunteer was given breathing instruction every 30 s and asked to return to the same respiratory level and maintain breath-hold for a couple of s, indicated by the short horizontal periods Depth (mm) Mean distance (mm) Volume (L) Heerink et al. European Radiology Experimental (2018) 2:22 Page 6 of 9 0 50 100 150 200 250 300 350 Time (s) Fig. 8 Respiratory level with feedback. Respiratory level of a volunteer (F4) in the prone position, with visual feedback. Red line is the initial level of breath-hold. Volunteer was asked to return to the same respiratory level every 30 s with the aid of visual biofeedback and maintain breath-hold for a couple of s, indicated by the short horizontal periods these numbers to the results of this study, one can con- respiratory motion from the study by Chen et al. [1] (up clude that even in the lower lung lobes, the biofeedback to 60 mm), this would result in a reproducibility of the system can potentially enable men and women to have a nodule position of within 1 mm. predictable consistent nodule position of well below Price et al. [6] have recently performed a clinical trial 0.5 mm. Even when considering nodules with extreme to assess the feasibility of using an in-house developed Position: Prone Without feedback With feedback F1 F2 F3 F4 M1 M2 M3 M4 Volunteer Fig. 9 Breath-hold error in prone position. Box plots of the error from initial breath-hold, without and with biofeedback, grouped by volunteer, in the prone position Mean distance (mm) Absolute error from inital breath-hold (mL) Heerink et al. European Radiology Experimental (2018) 2:22 Page 7 of 9 Position: Supine Without feedback With feedback F1 F2 F3 F4 M1 M2 M3 M4 Volunteer Fig. 10 Breath-hold in supine position. Box plots of the error from initial breath-hold, without and with biofeedback, grouped by volunteer, in the supine position optical surface tracking device to facilitate consistent study, the skin movement of the volunteers improved breath-hold during radiation therapy. They found that from 0.79 to 0.15 mm, when providing the feedback. patients were able to tolerate the feedback well and that Though these measurements cannot be directly com- they had a moderately improved reproducibility of skin pared, because they rely on technical factors as camera surface. They used traffic light colours to provide visual angle, our system should have a relatively higher rate of feedback to the patients and were able to reduce the improvement. mean amplitude of skin movement from 2.0 mm to The Kinect has the added benefit of being a generally 1.7 mm. In a previous healthy volunteer study [13], they available, low-cost system. Several groups have analysed achieved an improvement from 1.4 to 0.6 mm. In our the feasibility of using the Kinect camera to monitor re- spiratory motion for respiratory gated or four-dimensional CT-based continuous radiotherapy [3–5]. Though the Table 1 Absolute error from initial breath-hold at expiration results seem promising, to our knowledge, no clinical Median Q1–Q3 p value studies utilising the Kinect have been published yet. The Skin surface error without feedback (mm) 0.79 0.47–1.56 < 0.001 Kinect-based respiratory motion monitoring systems are Skin surface error feedback (mm) 0.15 0.06–0.30 mostly compared with the RPM Gating System (Varian E without feedback (mL) 147 76.8–276.8 < 0.001 overall Medical System, Palo Alto, CA, USA), a clinically available E with feedback (mL) 27.7 10.9–51.7 overall respiratory motion tracking system that utilised the move- ment of a marker box placed on the patient’s chest to gate E with feedback (mL) 35.5 14.3–52.3 0.081 men radiation therapy. This system is not suitable for interven- E with feedback (mL) 23.7 8.2–50.4 women tional procedures because the box has to be placed on E without feedback (mL) 124.6 55.7–259.7 0.012 supine disinfected skin and can easily be knocked out of place. As E without feedback (mL) 156.3 90.9–334.7 prone it only tracks the movement of a single marker, it would E with feedback (mL) 31.7 12.8–59.6 0.086 supine not be able to detect a change in breathing pattern either. E with feedback (mL) 22.3 7.7–47.0 prone During interventional procedures, patients are more likely E absolute error from initial breath-hold at functional residual capacity level to alter from thoracic to abdominal breathing, or vice Absolute error from inital breath-hold (mL) Heerink et al. European Radiology Experimental (2018) 2:22 Page 8 of 9 versa, rendering the tracking inaccurate. The value of skin glottis. Moreover, the spirometer data showed a signifi- surface motion tracking in combination with a tightly po- cant drift in the long-term measurements, even after sitioned vacuum mattress is that all respiratory movement rigorous (re)calibration of the spirometer, rendering can be visualised and thus be used as patient feedback. these long-term spirometry data unusable. Of note, the A change in breathing pattern is also a problem when spirometry measurements are not required for the sys- using abdominal/chest belts. These belts measure the tem to function in clinical practice. These were per- circumference of the patient’s chest or abdomen to formed only for validation of the system. provide patient feedback. Schoth et al. [14, 15] reported There are some limitations to this study. With a mean reduced intervention time and radiation exposure using age of 29 years and a mean body mass index of 22.2, the the IBC system (Mayo Clinic Medical Devices, USA) for volunteers were all young healthy adults, compared to the CT-guided lung biopsy while Carlson et al. [14, 15] potential target group. The breathing instructions resulted reported a reduction in targeting attempts using this belt in a breath-hold at FRC level. Obese patients generally in CT fluoroscopy-guided lung biopsy. However, in our breathe at a lower FRC [18] and patients suffering from experience these belts are cumbersome to setup and chronic obstructive pulmonary disease breathe at a higher unreliable. In a review of another bellows belt system FRC than healthy volunteers [19]. Although the relation (Philips Medical Systems, Eindhoven, The Netherlands), between chest wall motion and diaphragmatic excursion is Locklin et al. [2] demonstrated only a weak correlation approximately linear in healthy adults, this might not be between chest circumference and nodule position. true for patients [20]. Additionally, Harte et al. [21]have Another option that has been considered is the use of shown that patients with cystic fibrosis have a lower cor- spirometry to monitor lung volume. Tomiyama et al. [16] relation between chest wall movement and lung volume used a respiratory monitor to trigger an electric light bulb changes. Although differences in tracking accuracy are to as an indication of a similar level of breath-hold, resulting be expected between the volunteers and patients, the pro- in a high diagnostic accuracy (96%) in CT-guided biopsy cedure with biofeedback is feasible with low error rates of small (< 15 mm) lung nodules. The active breathing co- and easy to instruct to volunteers. ordinator system (Elekta Instrument AB, Stockholm, Patients might arguably have more difficulty in inter- Sweden) is a clinically available system used to actively preting the biofeedback and therefore in returning to monitor lung volume and suspend the patient’s breathing. their initial level of breath-hold every time. This does A valve closes and holds respiration at a certain level, to not have to lead to targeting errors per se, because the facilitate consisted tumour position for gated radiotherapy. operator will also see the biofeedback. He/she can there- From a practical standpoint, spirometry seems less suited fore keep on instructing the patient until the patient for interventional procedures, because it prevents com- manages to hold his breath is at the level the CT scan munication from patient to physician. If any breath were was acquired, before proceeding with needle manipula- to escape from the mouthpiece, the procedure would no tion. It should also be considered that patients who are longer be reliable. difficult to instruct will have more difficulty to return to As an alternative to CT-guided lung biopsy, CT fluor- a consistent level of breath-hold with only breathing oscopy can be used to target lung lesions. The advantage instructions, so these patients might benefit even more of CT fluoroscopy is the real-time feedback it provides from the feedback system. so the radiologist can verify the lesions’ position and im- Often, in interventional procedures parts of the thorax mediately place the biopsy needle. It has been reported and abdomen have to be covered with sterile drapes. with similar diagnostic accuracy and lower complication Movement of these drapes would result in an error in the rates, compared to CT-guided lung biopsy [17]. How- depth measurements. This can be prevented by either ever, both patient and radiologist experience a higher ex- using a drape with a large hole, and disinfecting a larger posure to radiation, which is a significant disadvantage. surface of the skin, or by using surgical incise drape with Additionally, depending on the CT scanner’s gantry tilt, adhesive backing. This facilitates a larger ROI to be the biopsy needle path is limited to in-plane approaches. selected, without having to include loose fitting drapes. During the biofeedback evaluation, the volunteers In conclusion, we presented a method to provide were not breathing into a spirometer. Instead, the patients with visual biofeedback of their respiratory level, spirometer was used during validation measurements to enable them to return to a consistent level of beforehand, to determine a conversion factor from chest breath-hold during image-guided interventions. The depth height to lung volume. When we attempted to use the measurements have proven to be an accurate measure of spirometer during the feedback evaluation, some volun- lung volume and the visual biofeedback enabled healthy teers demonstrated difficulty in maintaining breath-hold volunteers to return to 28 mL of their initial breath-hold because the mount piece prevented them from closing at expiratory level, corresponding to an estimated target their mouth, as if they were not able to close their position reproducibility of < 0.5 mm. If implemented in an Heerink et al. European Radiology Experimental (2018) 2:22 Page 9 of 9 image-guided intervention suite, it has the potential to 8. Holland AE, Goldfarb JW, Edelman RR (1998) Diaphragmatic and cardiac motion during suspended breathing: preliminary experience and prevent targeting errors caused by respiratory motion and implications for breath-hold MR imaging. Radiology 209:483–489 thereby to increase targeting accuracy. 9. Lanphier EH, Rahn H (1963) Alveolar gas exchange during breath holding with air. J Appl Physiol 18:478–482 Abbreviations 10. Lens E, Gurney-Champion OJ, Tekelenburg DR et al (2016) Abdominal organ C: Conversion factor; CT: Computed tomography; D: Difference; E: Absolute motion during inhalation and exhalation breath-holds: pancreatic motion at error; FRC: Functional residual capacity; IQR: Interquartile range; ROI: Region different lung volumes compared. Radiother On of interest 11. Manhire A, Charig M, Clelland C et al (2003) Guidelines for radiologically guided lung biopsy. Thorax 58:920–936 Availability of data and materials 12. Barret KE, Ganong WF (2012) Ganong’s review of medical physiology, 25th Data are available online in the Open Science Framework data repository edn. McGraw-Hill Medical, New York (https://osf.io/5ysfv/). 13. Parkhurst JM, Price GJ, Sharrock PJ, Jackson ASN, Stratford J, Moore CJ (2013) Self-Management of Patient Body Position, pose, and motion using wide-field, Funding real-time optical measurement feedback: results of a volunteer study. Int J This study was funded by Samenwerkingsverband Noord Nederland. Radiat Oncol Biol Phys 87:904–910 14. Schoth F, Plumhans C, Kraemer N et al (2010) Evaluation of an interactive Acknowledgments breath-hold control system in CT-guided lung biopsy. Rofo 182:507–511 Not applicable 15. Carlson SK, Felmlee JP, Bender CE et al (2005) CT fluoroscopy–guided biopsy of the lung or upper abdomen with a breath-hold monitoring and Authors’ contributions feedback system: a prospective randomized controlled clinical trial. WJH designed and conducted the study. WJH performed data analysis. WJH, Radiology 237:701–708 RV, PvO and MO performed data interpretation. WJH wrote the paper and all 16. Tomiyama N, Mihara N, Maeda M et al (2000) CT-guided needle biopsy of authors provided critical revisions and gave final approval. small pulmonary nodules: value of respiratory gating. Radiology 217:907–910 17. Kim GR, Hur J, Lee SM et al (2011) CT fluoroscopy-guided lung biopsy versus Ethics approval and consent to participate conventional CT-guided lung biopsy: a prospective controlled study to assess The study was approved by the local ethics committee and written informed radiation doses and diagnostic performance. Eur Radiol 21:232–239 consent was waived. 18. Parameswaran K, Todd DC, Soth M (2006) Altered respiratory physiology in obesity. Can Respir J 13:203–210 Consent for publication 19. Barnes PJ, Drazen JM, Rennard SI, Thomson NC (2009) Asthma and COPD Consent for publication has been provided by the volunteer pictured in basic mechanisms and clinical management. Elsevier Ltd, San Diego Fig. 2 and by the volunteer whose feedback is presented in Figs. 5 and 6. 20. Wang H-K, Lu T-W, Liing R-J, Shih TT-F, Chen S-C, Lin K-H (2009) Relationship between Chest Wall motion and diaphragmatic excursion in Competing interests healthy adults in supine position. J Formos Med Assoc 108:577–586 The authors declare that they have no competing interests. 21. Harte JM, Golby CK, Acosta J et al (2016) Chest wall motion analysis in healthy volunteers and adults with cystic fibrosis using a novel Kinect-based motion tracking system. Med Biol Eng Comput 54:1631–1640 Publisher’sNote Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Author details Center for Medical Imaging – North East Netherlands, University of Groningen, Groningen, The Netherlands. Department of Radiology, University of Groningen, University Medical Center Groningen, Groningen, The Netherlands. Department of Pulmonary Diseases, University of Groningen, University Medical Center Groningen, Groningen, The Netherlands. Received: 23 February 2018 Accepted: 29 May 2018 References 1. Chen A, Pastis N, Furukawa B, Silvestri GA (2015) The effect of respiratory motion on pulmonary nodule location during electromagnetic navigation bronchoscopy. Chest 147:1275–1281 2. Locklin JK, Yanof J, Luk A, Varro Z, Patriciu A, Wood BJ (2007) Respiratory biofeedback during CT-guided procedures. J Vasc Interv Radiol 18:749–755 3. Tahavori F, Alnowami M, Wells K (2014) Marker-less respiratory motion modeling using the Microsoft Kinect for windows. SPIE Medical Imaging 1–10 4. Lim SH, Golkar E, Rahni AAA (2014) Respiratory motion tracking using the Kinect camera. IEEE Conference on Biomedical Engineering and Sciences (IECBES). https://doi.org/10.1109/IECBES.2014.7047619 5. Ortmüller J, Gauer T, Wilms M, Handels H, Werner R (2015) Respiratory surface motion measurement by Microsoft Kinect. Current Directions in Biomedical Engineering 1:270–273 6. Price GJ, Faivre-Finn C, Stratford J et al (2017) Results from a clinical trial evaluating the efficacy of real-time body surface visual feedback in reducing patient motion during lung cancer radiotherapy. Acta Oncol 57:211–218 7. Liu HH, Koch N, Starkschall G et al (2004) Evaluation of internal lung motion for respiratory-gated radiotherapy using MRI: part II—margin reduction of internal target volume. Int J Radiat Oncol Biol Phys. 60:1473–1483

Journal

European Radiology Experimental – Springer Journals

Published: Sep 5, 2018

There are no references for this article.