비영리 - S-Space - 서울대학교

전체 글

(1)저작자표시-비영리-변경금지 2.0 대한민국 이용자는 아래의 조건을 따르는 경우에 한하여 자유롭게 l. 이 저작물을 복제, 배포, 전송, 전시, 공연 및 방송할 수 있습니다.. 다음과 같은 조건을 따라야 합니다:. 저작자표시. 귀하는 원저작자를 표시하여야 합니다.. 비영리. 귀하는 이 저작물을 영리 목적으로 이용할 수 없습니다.. 변경금지. 귀하는 이 저작물을 개작, 변형 또는 가공할 수 없습니다.. l l. 귀하는, 이 저작물의 재이용이나 배포의 경우, 이 저작물에 적용된 이용허락조건 을 명확하게 나타내어야 합니다. 저작권자로부터 별도의 허가를 받으면 이러한 조건들은 적용되지 않습니다.. 저작권법에 따른 이용자의 권리는 위의 내용에 의하여 영향을 받지 않습니다. 이것은 이용허락규약(Legal Code)을 이해하기 쉽게 요약한 것입니다. Disclaimer. (2) 의학박사 학위논문. Continuous pressure measurement and serial micro–computed tomography analysis during injection laryngoplasty. 성대 주입술 중 지속적 압력 측정 및 순차적 마이크로 단층 촬영을 이용한 분석. 2021년 2월. 서울대학교 대학원 의학과 이비인후과학 전공 김 민 수. I. (3) 의학박사 학위논문. Continuous pressure measurement and serial micro–computed tomography analysis during injection laryngoplasty. 성대 주입술 중 지속적 압력 측정 및 순차적 마이크로 단층 촬영을 이용한 분석. 2021년 2월. 서울대학교 대학원 의학과 이비인후과학 전공 김 민 수. III. (4) (5) Abstract Continuous pressure measurement and serial micro–computed tomography analysis during injection laryngoplasty Min-Su Kim Department of Medicine and Major in Otorhinolaryngology-Head and Neck Surgery The Graduate School Seoul National University. Injection laryngoplasty (IL) has been used to treat various types of glottal insufficiency. The precise volume and location of the injected materials impact the outcomes. However, exactly how increasing volumes of material are distributed is unknown. In fact, the amount of IL material required to medialize a vocal cord tends to be determined empirically. Thus, the goal of this study was to investigate the pattern of IL material distribution by checking serial micro–computed tomography (MCT) and pressure changes during ILs. This experimental study used 10 excised canine larynges. Experimental devices included the IL syringe, pressure sensor, infusion pump, fixed frame, and monitoring system. We injected calcium hydroxyapatite in the thyroarytenoid muscle; whenever 0.1 mL of material was injected, we obtained an MCT scan while simultaneously measuring the pressure. After the experiments, we performed histologic analyses. MCT analyses showed that materials initially expanded centrifugally and then expanded in all directions within the muscle. The pressure initially increased rapidly but then remained relatively constant until the point at which the materials V. (6) expanded in multiple directions. Histologic analyses showed that the IL material tended to expand within the epimysium of the thyroarytenoid muscle. However, in some cases, the MCT revealed that there were leakages to the surrounding space with a corresponding pressure drop. If the IL material passes through the epimysium, leakage can occur in the surrounding space, which can account for the reduction in resistance during ILs. ---------------------------------------------------------------------------------------------------------------Keywords: Dogs, Larynx, Medialization Laryngoplasty, Micro CT, Pressure Student Number: 2016-30584. VI. (7) TABLE OF CONTENTS. Abstract -------------------------------------------------------------------------------------------------- VI Table of Contents ------------------------------------------------------------------------------------ VIII List of Figures ------------------------------------------------------------------------------------------ IX. Introduction ----------------------------- ----------------------------------------------------------------- 1 Materials & Methods ------------------------------------------------------------------------------------ 2 Results ----------------------------------------------------------------------------------------------------- 9 Discussion ----------------------------------------------------------------------------------------------- 12 Conclusions --------------------------------------------------------------------------------------------- 16 References ----------------------------------------------------------------------------------------------- 17 Korean Abstract ---------------------------------------------------------------------------------------- 19. VII. (8) LIST OF FIGURES. Figure 1. An experimental cut of the canine larynx ------------------------------------------------- 2 Figure 2. Infusion pump -------------------------------------------------------------------------------- 3 Figure 3. Control box and infusion pump header ---------------------------------------------------- 4 Figure 4. Pressure sensor ------------------------------------------------------------------------------- 4 Figure 5. Pressure monitoring -------------------------------------------------------------------------- 5 Figure 6. Device to fix the canine larynx and infusion pump header in the micro CT tray ---5,6 Figure 7. Fixation device design ---------------------------------------------------------------------6,7 Figure 8. Schematic figure showing the experimental setup --------------------------------------- 7 Figure 9. Serial view of micro–computed tomography -------------------------------------------- 10 Figure 10. Analysis of the situation before leakage to the surrounding space ------------------ 11 Figure 11. Serial view of micro–computed tomography after paraglottic leak occurred ----- 11 Figure 12. Analysis of the situation after leakage to the surrounding space -------------------- 12. VIII. (9) Introduction Injection laryngoplasty (IL) has been used worldwide to treat various types of glottal insufficiency. It is known to be effective for preventing aspiration pneumonia and improving the voice in glottal insufficiency, such as in unilateral vocal paralysis [1]. Many types of injection materials have been developed and selected for different purposes, with accompanying reports on these materials’ effectiveness and safety. Although IL is now widely performed with many materials, the techniques used are highly variable [2-4]. Outcomes after ILs rely on the precise volume and location of the injected material; the volume and location can be challenging to control, especially for beginners, because of the pressure required to inject a viscous substance through a small needle. In fact, there are some unresolved questions for laryngologists. First, information on how the material is distributed within the larynx during an IL is limited. This information is needed to make decisions about where to place the injection syringe needle and how much material should be injected during an IL. Second, when the injection volume is low, many laryngologists find that the material disappears after a short time. However, to our knowledge, the redistribution pattern of the injected material has rarely been investigated scientifically. Third, the force required to push an injection syringe plunger initially increases as the amount of material injected increases; then, at some point, it is possible to inject the material with less force. This study was developed to solve these questions that have arisen from clinical experience during ILs. This study is a preliminary report on experimental tools that simultaneously measure both the distribution of the injected material and the change in 1. (10) injection pressure, while assuming the injection force needed to push the plunger of the injection syringe during an IL.. Materials and Methods This study was approved by the Institutional Animal Care and Use Committee of the Seoul National University Hospital (number 16-0208-S1A0) and was conducted in strict compliance with its requirements and determinations. A total of 16 larynges were excised from 16 Beagle dogs. The larynges were extracted from the carcasses after the completion of another canine experiment for a wearable cardiac pacemaker, which was also approved by the Institutional Animal Care and Use Committee (number 15-0313-S1A0) and was conducted through a specialized experimental animal company (Orient Bio, Seongnam, Republic of Korea). Of the 16 canine larynges, 6 were used in the pilot study for the development of the infusion pump and fixation device used in the micro–computed tomography (MCT). The other 10 canine larynges were used in the main experiment. Every canine larynges was refined like supraglottic laryngectomy to enter a limited space in MCT and to easily insert the injection needle to the thyroarytenoid muscle (Fig 1).. 2. (11) Figure 1. An experimental cut of the canine larynx. Preparation of Experimental Devices ILs were performed in the excised canine larynges using an infusion pump (NanoJet, Chemyx, Stafford, TX, USA) at a constant injection flow rate (Fig 2).. Figure 2. Infusion pump The infusion pump comprised an infusion pump header and a control box. The infusion pump header was responsible for pushing the plunger of the injection syringe and was located inside the MCT in order to continuously measure pressure during the IL. It was connected to the control box through wires outside of the MCT. The control box set up the flow rate and monitored the status of the injection (Fig 3).. 3. (12) Figure 3. Control box and infusion pump header The pressure sensor (ELVEFLOW, Paris, France) was located between the needle and the syringe, and was connected to the laptop through wires (RS232 cables) (Fig 4).. Figure 4. Pressure sensor Continuous pressure was measured during the experiments. Analog pressure data were converted to digital data using a sensor reader between the pressure sensor and the laptop. We analyzed the collected pressure data during the experiments in real time using ELVEFLOW 4. (13) software (Fig 5).. Figure 5. Pressure monitoring The pressure data shown in the results are the absolute pressure minus the atmospheric pressure. To prevent them from shaking, we used a device to fix the canine larynx and infusion pump header in the MCT tray (Fig 6A and B).. 5. (14) Figure 6. Device to fix the canine larynx and infusion pump header in the micro CT tray, (A) Frontal view, (B) Superior view This device was constructed using three-dimensional (3D) printing and designed with Solidworks software (Dassault Systems, Vélizy-Villacoublay, France) (Fig 7A and B).. 6. (15) Figure 7. Fixation device design, (A) Infusion pump fixation device, (B) Canine larynx fixation device The designed models were converted to stereolithography files and imported into Cubicreator 3D printing system software for prototyping (Cubicon Single Plus, Seongnam, Republic of Korea). All of this work was conducted through collaboration with 2 authors who are mechanical engineers (SA and SJ) (Fig 8).. 7. (16) Figure 8. Schematic figure showing the experimental setup. Experimental Protocol This experiment design is comparable to a transoral-style IL. The infusion pump was connected to the syringe for the IL (Fig 1). We used calcium hydroxyapatite (CAHA; Radiesse Voice, Merz Aesthetics, Frankfurt, Germany) as the injection material in the experiments, which were all performed by the same author (M-SK). Using a 25-gauge, 1.5inch needle, the author injected the CAHA into the thyroarytenoid (TA) muscles of the excised canine larynges, which were in the fixation device. The author injected the CAHA into the middle portion of each TA muscle. The needle was inserted into the lateral aspect of the vocal ligament. The CAHA was injected continuously into the TA muscle under a constant flow rate by the syringe connected to the infusion pump. The MCT visualization experiment was performed while measuring the pressure. MCT scans were taken whenever 0.1 mL of CAHA was injected, and the pressure was measured simultaneously in real time.. MCT and Histological Analysis An MCT scanner (NFR Polaris-G90; NanoFocusRay, Jeonju, Republic of Korea) was used to identify the injected material in the cadaveric canine larynx model. There were 600 MCT images, which were taken at settings of 65 kv, 60 µA, and 500 ms with a resolution of 512 × 512 in Digital Imaging and Communication in Medicine format. We serially analyzed the 8. (17) diffusion pattern of the injected material. Specifically, the MCT recorded the anatomic site and surrounding area whenever 0.1 mL of material was injected. After each experiment, the canine larynges were fixed in formaldehyde, decalcified with ethylenediaminetetraacetic acid, cut into 0.5-cm slices, and embedded in paraffin blocks. Sections were stained for the histological analysis using hematoxylin and eosin.. Results The interval between 0.1-mL injections ranged from 67 to 123 seconds (median, 95 seconds). The time taken for injection ranged from 7.0 to 7.5 seconds (median, 7.2 seconds). The MCT analysis showed that in the early stages (0.1–0.2 mL) of IL, the CAHA proceeded in a centrifugal and spherical expansion around the needle tip (Fig 9A–C). As the IL continued, the spherical expansion resulted in a medial bulging of the vocal fold but was blocked by the vocal ligament and thyroid cartilage. As the IL proceeded, the CAHA began to expand within the TA muscle, moving forward, backward, up, and down from the injection point. The downward expansion stopped at the cricothyroid space, and the superior expansion caused bulging to the ventricle (Fig 9D–F).. 9. (18) Fig 9. Serial view of micro–computed tomography. (A) Axial view at initial injection. (B) Coronal view at initial injection. (C) Sagittal view at initial injection. (D) Axial view before leakage. (E) Coronal view before leakage. (F) Sagittal view site before leakage.. Histologic examinations showed that the CAHA was dilated within the TA muscles. It expanded in an oval pattern within the muscle because it was blocked by the thyroid cartilage and the vocal ligament (Fig 10A). The pressure initially increased rapidly but then remained relatively constant (Fig 10B). If the CAHA did not leak, 1.3 mL was injected.. 10. (19) Fig 10. Analysis of the situation before leakage to the surrounding space. (A) Histologic analysis. Bar: 1 mm. (B) Pressure measurement.. All 10 injection experiments showed the same pattern as described above before leaks occurred in some cases. However, CAHA leaked to the surrounding spaces in some cases during the IL. In 1 case, there was a leak to the paraglottic space (Fig 11). In other cases, we observed leaks to the airway through the hole in which the injection needle was inserted.. 11. (20) Fig 11. Serial view of micro–computed tomography after paraglottic leak occurred. (A) Axial view. (B) Coronal view. (C) Sagittal view.. In those cases, histologic examinations showed that the CAHA leaked when there was a TA muscle rupture (Fig 12A). The pressure initially increased rapidly, remained relatively constant, and then dropped rapidly after leakage (Fig 12B). When pressure dropped due to a leak, 0.9 to 1.1 mL of CAHA had been injected.. Fig 12. Analysis of the situation after leakage to the surrounding space. (A) Histologic analysis. Bar: 1 mm. (B) Pressure measurement.. Discussion In a previous study, the safety of IL with CAHA was reported in nearly 1,000 cases [5]. 12. (21) Among the various materials used for IL, CAHA is widely used for long-term vocal fold augmentation. CAHA is easily visualized in computed tomography (CT) because the calcified material is radiopaque. We previously reported that CAHA can be adequately observed in CT as a bright whitish mass in the glottic and paraglottic space [6]. Therefore, CAHA is the best material to use when performing a visual analysis of the distribution pattern of injected material. However, it is ethically problematic to conduct serial CT while simultaneously performing an IL in the human body. Therefore, we performed this study using excised canine larynges instead. The canine larynx is widely used as an appropriate model for studying human laryngeal physiology, because the size and histological properties of the canine and human larynxes are similar [3, 7-9]. In addition, in our experiment the volume and velocity of the injection were controlled by a modified syringe pump. The force required to push a plunger can vary based on the size of the syringe and needle. The force is calculated by multiplying the pressure by the area. The area of the syringe containing the injection material is constant. Therefore, we measured the pressure changes in the syringe, which represent the changes of force. The most problematic part of this study was the simultaneous installation of the excised canine larynx, fixation device, syringe, pressure sensor, and syringe pump inside the limited space of the MCT tray in order to record data from the devices in real time. A transcutaneous-style IL experiment could not be simulated because of the short height of the MCT tray space, as the syringe pump with the IL syringe had to be mounted diagonally. Thus, because of the narrow MCT tray space, the simulation used a transoral-style IL experiment, in which the needle passes horizontally through the upper part of the vocal cords. From the results of this experiment, we can interpret information about the distribution 13. (22) of injection material. First, in the early stages of an IL, the injection material expanded spherically in a centrifugal direction because the IL material initially pushed outward within the TA muscle. However, as the amount of IL material increased, it started to expand in an oval pattern along the longitudinal axis of the vocal cord, where the resistance is low; this is probably because the vocal ligament, conus elasticus, and thyroid cartilage acted as barriers. A histologic analysis showed that the injection material stayed in the TA muscle space in most cases. However, when the IL material exceeded a certain amount, it leaked to the paraglottic space from the TA muscle. We assume that the intramuscular pressure exceeded the elastic limit of the epimysium of the TA muscle at this point, which resulted in a rupture at its weakest area that, in turn, allowed the material to leak into the surrounding space. When the leakage to the surrounding space occurred, an additional injection might not affect the medialization of the vocal cord. Our histological analysis showed how CAHA leaked out of the TA muscle and into the paraglottic space. We assume that once CAHA leaks to the paraglottic space, it can be redistributed to the base of the pyriform sinus, aryepiglottic fold, or postcricoid space, which are connected posteriorly by loose areolar tissue. This assumption explains why we sometimes find a lump of injection material in the posterior cricoid space during the arytenoid adduction procedure after a CAHA injection. Some studies have shown similar results. Mau et al performed a transoral-style IL experiment with CAHA in human cadaveric larynges, checking the CT scans both before and after the IL. The results showed that the CAHA tended to have a relatively compact appearance on CT and was distributed along the length of the vocal fold when injected medially. By contrast, in 20% of female cadaver larynges (3/15 cases), CAHA leaked through the cricothyroid space outside of the laryngeal framework when injected laterally[10]. However, our results did not show leakage 14. (23) to the cricothyroid space. The difference in these results is believed to be due to the use of the relatively larger canine larynx with a lateral injection, as compared to a human female larynx without a lateral injection. In addition, the loss of the medialization effect after a short time can be attributed to the redistribution of injection material to the surrounding spaces during the IL. Second, if the above hypothesis is correct, it is possible that the abrupt reduction of resistance to the syringe during the IL procedure is a result of an epimysium rupture. The resistance felt by a laryngologist during an IL may have various causes. The most common cause of resistance is clogging of soft tissue or cartilage in the needle when the operator begins to push the plunger. However, the resistance decreases once the IL proceeds. Resistance caused by needles approaching the arytenoid cartilage or the vocal ligament in close proximity is also common. In this situation, the resistance can decrease if the needle is repositioned. However, in the present study we experienced a decrease in resistance during injection, unlike in the scenarios described above. Once the resistance decreases, the remaining material is injected very easily. This phenomenon may be represented as a pressure drop due to an epimysium rupture. In our experiment, some cases showed a sudden drop in pressure, which might support this hypothesis. Thus, our findings suggest that the IL material escapes somewhere, moving away from the TA muscle. This study has some strengths. First, this was a nearly real-time analysis of IL, which has rarely been conducted. Second, we accurately and scientifically measured injection pressures. Third, we analyzed MCT serially and variously (i.e., in the axial, coronal, and sagittal views). This visual analysis enabled a profound understanding of the distribution patterns of the injection material. 15. (24) The present study also has some limitations. First, we did not simulate a transcutaneousstyle experiment. Future transcutaneous-style experiments may demonstrate a pattern of injection material distribution that differs from the current results. Second, because of the time needed for MCT scanning, a step-by-step injection method was used, which can result in a different outcome than an actual clinical continuous injection. Third, we did not investigate various IL materials, such as hyaluronic acid or carboxymethylcellulose. Each material has unique viscosity and resorption characteristics [11]. Therefore, results with different IL materials may differ from the findings of this study. Finally, the canine larynx is different from the human larynx, although it is considered to be a good experimental alternative to the human larynx. In addition, we used frozen and thawed canine larynges, which may react differently from non-frozen canine larynges. In the future, we will conduct a more controlled and improved experiment. It is possible to verify our hypothesis by increasing the number of experimental objects and creating a transcutaneous-style model to observe the distributed patterns. Additionally, using pressure sensors that can be connected to IL syringes in real clinical situations could help to conduct the future research on adequate volume for IL.. Conclusions The IL material expands within the epimysium of the TA muscle, and its range is limited in the vocal ligament, conus elasticus, and thyroid cartilage. However, if the injection material passes through the epimysium, leakage can occur in the surrounding space, which can account for the reduction in resistance during an IL.. 16. (25) References 1.. Rosen CA, Amin MR, Sulica L, Simpson CB, Merati AL, Courey MS, et al. Advances. in office-based diagnosis and treatment in laryngology. Laryngoscope. 2009;119(Suppl 2):S185-212. doi:10.1002/lary.20712. 2.. Rosen CA, Thekdi AA. Vocal fold augmentation with injectable calcium. hydroxylapatite:. short-term. results.. J. Voice.. 2004;18(3):387-391.. doi:10.1016/j.jvoice.2004.02.001. 3.. Lim JY, Kim HS, Kim YH, Kim KM, Choi HS. PMMA (polymethylmetacrylate). microspheres and stabilized hyaluronic acid as an injection laryngoplasty material for the treatment of glottal insufficiency: in vivo canine study. Eur Arch Otorhinolaryngol. 2008;265(3):321-326. doi:10.1007/s00405-007-0458-y. 4.. Lodder WL, Dikkers FG. Comparison of voice outcome after vocal fold augmentation. with. fat. or. calcium. hydroxylapatite.. Laryngoscope.. 2015;125(5):1161-1165.. doi:10.1002/lary.25104. 5.. Lee M, Lee DY, Kwon TK. Safety of office-based percutaneous injection. laryngoplasty with calcium hydroxylapatite. Laryngoscope. 2019;129(10):2361-2365. doi:10.1002/lary.27861. 6.. Kwon TK, Lee JE, Cha WJ, Song CM, Sung MW, Kim KH. Serial computed. tomographic and positron emission tomographic properties of injection material used for vocal fold augmentation. Laryngoscope. 2013;123(10):2491-2496. doi:10.1002/lary.24017. 7.. Chhetri DK, Jahan-Parwar B, Hart SD, Bhuta SM, Berke GS. Injection laryngoplasty. with calcium hydroxylapatite gel implant in an in vivo canine model. Ann Otol Rhinol Laryngol. 2004;113(4):259-264. doi:10.1177/000348940411300402. 17. (26) 8. an. Noordzij JP, Perrault DF, Jr., Woo P. Biomechanics of arytenoid adduction surgery in ex. vivo. canine. model. Ann. Otol. Rhinol. Laryngol.. 1998;107(6):454-461.. doi:10.1177/000348949810700602. 9.. Hoffman MR, Witt RE, Chapin WJ, McCulloch TM, Jiang JJ. Multiparameter. comparison of injection laryngoplasty, medialization laryngoplasty, and arytenoid adduction in an excised larynx model. Laryngoscope. 2010;120(4):769-776. doi:10.1002/lary.20830. 10.. Mau T, Courey MS. Influence of gender and injection site on vocal fold augmentation.. Otolaryngol Head Neck Surg. 2008;138(2):221-225. doi:10.1016/j.otohns.2007.10.028. 11.. Mahboubi H, Mohraz A, Verma SP. Evaluation of heating and shearing on the. viscoelastic properties of calcium hydroxyapatite used in injection laryngoplasty. Otolaryngol Head Neck Surg. 2016;154(3):498-501. doi:10.1177/0194599815625206.. 18. (27) 요약 (국문초록) 성대 주입술 중 지속적 압력 측정 및 순차적 마이크로 단층 촬영을 이용한 분석. 서울대학교 대학원 의학과 이비인후과학 전공 김민수. 성대 주입술은 다양한 유형의 성문 폐쇄 부전을 치료하기 위해 사용되어 왔다. 주입 물질의 정확한 부피와 위치는 음성 결과에 영향을 미친다. 그러나, 주입 물 질의 양이 증가할수록 어떻게 분포되는지는 정확히 알려지지 않았다. 실제로 성 대의 내측화에 필요한 주입 물질의 양은 경험적으로 결정되는 경향이 있었다. 따 라서 본 연구의 목적은 성대 주입술 중 순차적 마이크로 단층 촬영과 지속적 압 력 변화를 확인함으로써 성대 주입 물질의 분포 패턴을 조사하는 것이다. 이 실 험 연구에서는 10개의 절제된 개 사체 후두를 사용했다. 실험 장치로는 성대 주 입술 주사기, 압력 센서, 주입 펌프, 고정 프레임 및 모니터링 시스템들이 포함되 었다. 우리는 주입 물질로 갑상피열근에 수산화인회석 칼슘을 주입했다. 0.1 mL 의 물질을 주입할 때마다 압력을 측정하면서 마이크로 단층 촬영을 시행했다. 실 험이 종료된 후, 우리는 주입이 완료된 개 사체 후두들에 대한 조직학적 분석을. 19. (28) 수행했다. 마이크로 단층 촬영 분석에서 주입 물질은 주입 초기에는 원심형으로 팽창하다 이후에 근육 내 모든 방향으로 팽창했다. 압력은 주입 초기에는 급격히 증가했지만, 이후에는 주입 물질이 근육내에서 다양한 방향으로 팽창하는 동안 비교적 일정하게 유지되었다. 조직학적 분석에 따르면 성대 주입 물질은 갑상피 열근의 외막내에서 팽창하는 경향이 있었다. 하지만 일부의 경우에서는 마이크로 단층 촬영상 주입 물질이 주변 공간으로 유출되면서 동시에 발생하는 압력의 강 하가 있었다. 만약 성대 주입 물질이 근육 외막을 뚫고 통과되는 경우에는 주변 공간으로 유출이 발생할 수 있고 이것은 우리가 임상적으로 성대 주입술시 종종 느끼는 저항의 감소 현상을 설명할 수 있을 것이다. --------------------------------------------주요어: 개, 후두, 내측화 성대성형술, 마이크로 단층 촬영, 압력 학번: 2016-30584. 20. (29) 감사의 글 많은 어려움 끝에 얻은 서울대학교 의과대학 이비인후과 의학박사 취득이라 감사 한 분이 너무나 많아 글을 남깁니다. 사랑하는 가족들인 아내 김지인, 아들 김주 원, 그리고 한달 후 태어날 딸에게 그동안 너무 감사했고 많이 부족해서 미안했 다는 말을 전합니다. 이 논문을 구현하는데 도움을 주신 한국산업기술대 기계공 학과 진송완 교수님께도 깊이 감사드립니다. 교수님이 아니셨다면 공학적 지식이 전무한 저로서는 이 논문은 시작조차 할 수 없었다고 생각합니다. 불쑥 어느 날 연락드리고 찾아갔음에도 진지하게 도움을 주신 점 감사드립니다. 또한 한국산업 기술대 기계공학과 안성민 학생에게도 감사드립니다. 2021년부터는 서울대학교 공과대학원에도 진학예정이라 들었습니다. 안성민 학생이 아니었다면 그 무거운 기계들을 갖고 실험을 혼자는 할 수 없었다고 생각합니다. 마지막으로 저의 석박사 지도 교수님인 권택균 교수님께 감사드립니다. 앞날이 불투명하고 불안해하 던 펠로우였던 저에게 유일하게 할 수 있다는 자신감을 주셨던 분이십니다. 현재 누구보다도 씩씩하게 한 분야의 전문가로서 발전을 위해 나아가는 저의 모습은 선생님이 없었다면 불가능했을 것입니다. 모든 분들께 진심으로 감사드립니다.. 21. (30)