Access the full text.
Sign up today, get DeepDyve free for 14 days.
Current Directions in Biomedical Engineering 2019;5(1):21-24 Ivan Maldonado*, Alfredo Illanes, Axel Boese, Marco Kalmar, Thomas Sühn and Michael Friebe Audio waves and its loss of energy in puncture needles https://doi.org/10.1515/cdbme-2019-0006 Introduction Abstract: The location of a puncture needle’s tip and the resistance of tissue against puncture are crucial information for clinicians during a percutaneous procedure. The tip location Percutaneous procedures have proven to be efficient and safe and needle alignment can be observed by image guidance. during the extraction of a sample tissue for a diagnosis or a Tactile information caused by tissue resistance to rupture, treatment that requires the administration of medication . allow clinicians the perception of structural changes during The main factors around the success and clinician’s puncture. Nevertheless, this sense is individual and subjective. acceptance are its minimal discomfort on the patients, low To improve percutaneous procedures, the implementation of cost, and precision. Last-mentioned, thanks to the transducers to enhance or complement the senses offer implementation of medical imaging techniques as ultrasound objective feedback to the user. Known approaches are e.g. or magnetic resonance. based on integrated force sensors. However, this is connected Initially, the success of a percutaneous procedure to higher device costs, sterilization and certification issues. A implementing an imaging technique was associated with the recent publication shows the implementation of an audio clinician’s skills and experience. Nevertheless, because its transducer clipped at the proximal end of the needle. This subjectivity, different researches in robotics , sensors sensor is capable of acquiring emitted sounds of the distal tip- , navigation , tracking  and enhancement tissue interaction that are transmitted over the needle structure. algorithms  have shown to be objective, useful for clinicians The interpretation of the measured audio signals is highly during a procedure and, most of them successfully depended on the transmission over the needle, the tissue and, implemented on real scenarios. In the particular case of the penetration depth. To evaluate the influence of these sensors, the majority of the applications remain on research parameters, this work implements a simplified experimental and do not reach an operation room because regulations that setup in a controlled environment with a minimum of noise make the approval of sensors extremely difficult. In most of and without micro tremors induced by clinician’s hands. A the cases, sensors are located on the tip or any part of the steel rod simulating a needle is inserted into pork meat of needle´s shaft in contact with tissue. different thickness. A controlled impact covering the needle’s A recent publication  demonstrated that an audio tip mimics tissue contact. The resulting signals are recorded transducer attached to the proximal out-patient end of a and analyzed for better understanding of the system. puncture needle is able to acquire relevant information generated from the interaction between the tip and tissue in the Keywords: Percutaneous procedure, audio transducer, distal part. Until now, the concept of an audio transducer needle sensor, parametrization, classification of tissue, attached to a needle can identify the transition between two characterization of tissue. different types of tissue. But it is unclear how the acquired signal is influenced by the interaction between the needle and soft tissue and the transmission of vibroacoustic waves. This work identifies parameters that help to understand the transfer function of the needle when it is located in two ______ different depths inside muscle pork tissue. The identification *Corresponding author: Ivan Maldonado, Institute of Medical of these parameters can be the first step for the characterization Technology, INKA, Otto-von-Guericke-University, Magdeburg, and classification of tissue. Germany, e-mail: firstname.lastname@example.org Alfredo Illanes, Marco Kalmar, Thomas Sühn, Axel Boese and Michael Friebe, Institute of Medical Technology, INKA, Otto-von- Guericke-University, Magdeburg, Germany. Open Access. © 2019 Ivan Maldonado, Alfredo Illanes, Marco Kalmar, Thomas Sühn, Axel Boese and Michael Friebe, published by De Gruyter. This work is licensed under the Creative Commons Attribution 4.0 License. I. Maldonado et al., Audio waves and its loss of energy in puncture needles — 22 Materials and Methods Design of experiment The implemented experiment reproduces a needle procedure when its tip is located in three different depths inside bio tissue. During puncture, the needle’s tip cuts fibre layers of the tissue. Due to the shape of the tip, this interaction causes longitudinal and radial forces that induce micro-vibrations . These vibrations travel along the rigid needle’s shaft. An audio transducer at the proximal end is capable of acquiring these vibrations as sound signals. The transmission of the micro-vibrations is expected to change when the depth of the needle increases. To prove that, an experimental setup was implemented. From a mechanical point of view, a needle in a doctor’s Figure 2: Experimental setup. A) Audio transducer attached to 3D holder with double tape. B) Steel rod’s tip must be free for impact. hand can be simplified as a free oscillating cantilever (see C) Impact actuator. The maximum tissue pad is 5mm Figure ). When the tip is in contact with tissue, this model changes to the simply supported beam. Increasing the a total length of 210 mm. 10 mm of the proximal end of the penetration depth reduces the free oscillating length of this rod were embedded to a 3D printed grip. An audio transducer beam and, increases the fixation or damping force on the distal (MEMS microphone SPH0645LM4H-B, Knowles, Illinois, end. USA) was fixed with double-sided tape with a centered hole When an impact is given to the structure, the system starts to the 3D printed grip. The tape acts as environmental noise to oscillate. Depending on the impact this oscillation is defined cancellation, in which the center hole allows sound waves to by a maximum amplitude and a damping constant of the travel to the microphone. (see Figure A). system. These parameters should influence the acquired audio A Raspberry Pi 3 (Raspberry Pi Foundation, Cambridge, signal. UK) computer stored and received all data acquired by the audio transducer via an I2S communication protocol. The implemented programming language to collect and store the Experimental Setup data was written in Python 2.7. An actuator was set to impact the needle’s tip in a A steel rod behaves as a simplified model of a medical controlled manner. It consists of a servomotor (Analog servo puncture needle (see Figure ). It was fixed horizontally and 033212, JAMARA, Aichstetten, Germany) attached to a 3D held from one end. The steel rod had a diameter of 1.5 mm and printed hammer with a cavity at the tip. The cavity was filled with 5 mm of pork meat (see Figure C) for simulating the breaking through tissue layers during impact. An Arduino device (Arduino Mega 2560, arduino.cc, Somerville, U.S.A.) controlled the servomotor via Raspberry Pi commands. For signal evaluation, a programming environment using MATLAB 2018b was created. Measurement Protocol Three different scenarios were considered. In the first scenario, the shaft of the rod was completely free; in the second and third scenario, the rod was inserted in 4 and 8 cm Figure 1: Simplified mechanical representation example. A) First scenario. Simple cantilever. B) Second scenario. Reduction of of pork meat respectively (see Figure and 2). The Python’s free oscillation in a length of 4 centimeters. C) Third scenario. software loaded at the Raspberry Pi device initialized the audio Reduction of free oscillation in a length of 8 centimeters. I. Maldonado et al., Audio waves and its loss of energy in puncture needles — 23 Signal Processing The main objective of the signal processing strategy was to compute significant parameters for characterizing the audio signal under different scenarios. For that, we assume that the signal resulting from the excitation impact can be parametrized using a simple damped sinusoidal model. For that, first, segments of 0.20 seconds were manually selected (see Figure 3a). This selection has as main objective to only take into account the signal segment corresponding to the needle response, avoiding influences of the impact per se or noise generated by the servomotor. Then, the influence of gravity was attenuated by trend removing using a multilevel one-dimensional discrete wavelet transformation with a Daubechies mother wavelet , (see Figure 3b). The resulting signal was then denoised using the same aforementioned wavelet transformation but with a higher frequency scale (see Figure 3c). The detrended and denoised signal was squared in order to estimate the instantaneous energy (see Figure 3d). Finally, the peaks of the energy signal higher than 10 % of the maximum instantaneous energy were detected and used as data to be fitted by the exponential model (see Figure 3e) ′ 𝑏 ∙𝑥 𝑦 ≈ 𝑎 ∙𝑒 (1) where 𝑎 is the point of maximum instantaneous energy, 𝑏 is the constant of decay and, 𝑦 represents the peaks of maximum Figure 3: Signal processing. Left to right and up to down: a) manually selection; b) remove of gravity, c) denoising, d) instantaneous energy consequence of the free oscillation instantaneous energy of signal and e) exponential fit procedure. induced in the system by simulating the breaking through of the tip. transducer recording and an internal timer. After 5 seconds, a command to a secondary logical program located at the Arduino initialized the actuator sequence that position the actuator twenty-five degrees from the tip, impact the needle’s tip and return to its original position in a time of 0.2 seconds. It is important to clarify that the tip of the rod remains free in every moment to be able to receive the impact (see Figure B) and, the tissue set inside the 3D printer cavity had the same properties than the one surrounding the shaft. Based on repeated experiments, it was possible to estimate that the actuator collided the needle’s tip with a force of 1.5 N. All actuator’s impacts were video recorded with a conventional Android smartphone at a resolution of FHD 1080p. The videos allow the visualization of the procedure for research and demonstration purposes. The audio signals were stored as Waveform Audio File Format (WAV). Each audio file had a duration of 10 seconds with a sampling frequency of 44100 Hz. In total, 60 audio Figure 4: Boxplot from the parameter constant of decay. Each group is clearly differentiated from each other. signals, 20 for each scenario and each of them with a single measure, were acquired using an audio transducer. I. Maldonado et al.,Audio waves and its loss of energy n i puncture needles — 24 Author Statement Results and Discussion The Federal Ministry of Education and Research (BMBF) financially supported this research in the context of the INKA The parameters of maximum instantaneous energy and Project (Grant Number 03IPT7100X) and CONACYT constant of decay demonstrated to vary in correlation with the Mexico. Authors state no conflict of interest. Informed consent depth of the steel rod inside the pork meat (Table 1). The has been obtained from all individuals included in this study. parameter of maximum instantaneous energy has a tendency to be more stable when the depth increase, opposite to the parameter constant of decay. The parameter of maximum References instantaneous energy can be associated with waves of resonance meanwhile the constant of decay to the time of  Vlastos, G., & Verkooijen, H. M. (2007). Minimally invasive approaches for diagnosis and treatment of early-stage breast energy dissipation. cancer. The oncologist, 12(1), 1-10. The parameter constant of decay shows a clear  Kaya, M., Senel, E., Ahmad, A., & Bebek, O. (2019). Visual aggrupation of the different scenarios based on how quickly needle tip tracking in 2D US guided robotic interventions. the energy dissipates (see Figure 4). It means that from the first Mechatronics, 57, 129-139.  Abayazid, M., Pacchierotti, C., Moreira, P., Alterovitz, R., point of maximum instantaneous energy, until the second peak Prattichizzo, D., & Misra, S. (2016). Experimental evaluation of maximum instantaneous energy, the time varies in of co‐manipulated ultrasound‐guided flexible needle steering. correlation with the tissue’s depth. The values of this The International Journal of Medical Robotics and Computer parameter do not overlap independently of their variance. Assisted Surgery, 12(2), 219-230.  Tiwana, M. I., Redmond, S. J., & Lovell, N. H. (2012). A review of tactile sensing technologies with applications in Table 1: Result of maximum instantaneous energy and constant of biomedical engineering. Sensors and Actuators A: physical, decay of 20 audio signals for each scenario. 179, 17-31.  Puangmali, P., Althoefer, K., Seneviratne, L. D., Murphy, D., & Dasgupta, P. (2008). State-of-the-art in force and tactile Maximum sensing for minimally invasive surgery. IEEE Sensors Const. Of Decay Instantaneous Energy Scenario Journal, 8(4), 371-381.  Zettinig, O., Frisch, B., Virga, S., Esposito, M., Rienmüller, Average Variance Average Variance A., Meyer, B., ... & Navab, N. (2017). 3D ultrasound registration-based visual servoing for neurosurgical Free 93.90 42.50 -4.87 0.38 navigation. International journal of computer assisted 4 cm meat 70.73 135.57 -15.05 12.32 radiology and surgery, 12(9), 1607-1619. 8 cm meat 97.03 5.70 -53.53 100.30  Xia, W., Noimark, S., Ourselin, S., West, S. J., Finlay, M. C., David, A. L., & Desjardins, A. E. (2017, September). Ultrasonic needle tracking with a fibre-optic ultrasound transmitter for guidance of minimally invasive fetal surgery. In International Conference on Medical Image Computing and Conclusions Computer-Assisted Intervention (pp. 637-645). Springer, Cham. The audio signals follow the behavior of the mechanical model  Mwikirize, C., Nosher, J. L., & Hacihaliloglu, I. (2018). Signal attenuation maps for needle enhancement and localization in and, the parameters approached by an exponential form (Eq. 2D ultrasound. International journal of computer assisted 1) allow the differentiation of three scenarios in interaction radiology and surgery, 13(3), 363-374. with the same tissue.  Illanes, A., Boese, A., Maldonado, I., Pashazadeh, A., The calculation of the parameters of maximum Schaufler, A., Navab, N., & Friebe, M. (2018). Novel clinical device tracking and tissue event characterization using instantaneous energy and constant of decay contain proximally placed audio signal acquisition and processing. information that indicates how fast the energy in the system Scientific reports, 8(1), 12070. dissipate based on the surrounding tissue. As future work, the  Barnett, A. C., Lee, Y. S., & Moore, J. Z. (2016). Fracture system must be exposed to different tissues and their mechanics model of needle cutting tissue. Journal of Manufacturing Science and Engineering, 138(1), 011005. parameters must be analyzed. Advanced signal processing  Abolhassani, N., Patel, R., & Moallem, M. (2007). Needle must be implemented to reduce ambient noise and micro insertion into soft tissue: A survey. Medical engineering & tremors induced by the clinicians. The upcoming analysis must physics, 29(4), 413-431 conclude if the information contained in the audio signals can  Bogusz, J., Klos, A., & Kosek, W. (2013). Wavelet be used for the extraction of useful features for a classification decomposition in the Earth’s gravity field investigation. Acta Geodynamica et Geomaterialia, 10(1), 169. algorithm.
Current Directions in Biomedical Engineering – de Gruyter
Published: Sep 1, 2019
Keywords: Percutaneous procedure; audio transducer; needle sensor; parametrization; classification of tissue; characterization of tissue
Access the full text.
Sign up today, get DeepDyve free for 14 days.