Graduate School of Advanced Science and Engineering, Waseda University

(1)

Graduate School of Advanced Science and Engineering, Waseda University

༤ ༤㻌 ኈ㻌 ㄽ㻌 ᩥ㻌 ᴫ㻌せ

Doctor Thesis Synopsis

ㄽ ᩥ 㢟 ┠

Thesis Theme

Development of an EMG+IMU Driven Musculoskeletal Model and its Applications in Medical Training

Assessment

➽㟁᝟ሗ࡜ጼໃ᝟ሗ࡟ࡶ࡜࡙ࡃ➽άື᥎ᐃ

ࣔࢹࣝࡢ㛤Ⓨ࡜ࡑࡢ་⒪ᡭᢏホ౯࡬ࡢᛂ⏝

⏦ ㄳ ⪅ (Applicant Name)

Luca BARTOLOMEO

ࣂࣝࢺ࣓ࣟ࢜ ࣝ࢝

⏕࿨⌮ᕤᏛᑓᨷ ࣂ࢖࣭࢜ࣟ࣎ࢸ࢕ࢡࢫ◊✲

2013 ᖺ 5 ᭶

(2)

No.1

The main objective of medical education is to guarantee that the graduating physicians are competent to practice in their selected field and that the preparation among them is independent of the medical school. An unbalance of preparation among medical doctor of different hospitals is translated directly into a difference of treatment for the patients.

In order to increase the homogeneity among the medical schools, many countries are recently adopting a competency-based training, in which the residents are evaluated on their acquired skills, eliminating the concept of years-in-training. Each trainee has the autonomy to proceed based on his or her capability, with some intermediate steps to provide evidence of the mastery of what is required. The evaluation of competency is assured by skilled impartial observers, judging who performs a procedure adequately and who does not.

However, there is a high risk of the assessors subjectively interpreting the results, with the outcome depending on background, skill, and experience of the assessors. This makes the evaluation highly variable from place to place, and from assessor to assessor. Additionally, it is difficult to secure the time of senior surgeons, because they are extremely busy for their own work and not all facilities can count on such skilled personnel. The increased complexity of modern medical technologies and equipment has made the assessment even more problematic. Definitely, without an objective evaluation of competency to practice, the homogeneity of preparation cannot be guaranteed.

Training devices, such as box trainers and virtual reality simulators, have already been developed to learn fundamental operative skills outside the operation room and to measure the performance during the training.

However, the science of objective evaluation of operative performance is still in its early stages. Currently, the objective skill evaluation systems on most of the training facilities have several limitations, which can be basically summarized into two main aspects:

1) In the cases of virtual reality or dedicated simulators, they provide only low-level skill scoring by using the execution time and instrument’s motion information for completing the training tasks; additionally they cannot be extended to other medical training systems.

2) Existing general objective skill evaluation methods use only motion analysis or only muscle force analysis, missing one of the two information. To date, there is no common methodology capable of integrating motion analysis and muscle force estimation to achieve objective skill evaluation in medical applications.

In this thesis, the road towards the realization of a common objective skill evaluation system for multiple medical applications is presented. The specific goal of this work is to develop a new integrated biomechanical model for the objective assessment of medical skills, and to provide it to the assessors for an impartial and exhaustive evaluation and coaching of the trainees. More specifically, the proposed system aims at answering the questions: how to discover and analyze objectively the trainees’ weak points, and how to overcome them in the best possible way. The advantage is that virtually all the physicians in all the hospitals in the world can have the same high level of training/evaluation as the top ones, with an evident benefit for the patients all around the world.

(3)

No.2 In this work the aims are achieved by:

1) developing a common skill evaluation system, which is independent from the training devices and can be adaptive for various medical training scenarios.

2) combining human motion tracking system and surface electromyography in an EMG-Motion model suitable for analyzing the biomechanics of trainee’s upper body during practical medical training.

3) verifying that the proposed skill evaluation system can be implemented in multiple medical applications for evaluating the surgical operative skills and assisting clinical practice.

At the end, this research demonstrates a general methodology for objectively evaluating medical operative skills. The model can be used not only for regular operative skill trainings, but can be also further implemented for clinical practice and other medical scenarios.

The validity of the resulting skill evaluation system is verified in the field of laparoscopy and endotracheal intubation, covering applications in both operating room and emergency room. The evaluation for surgery is based on the assumption that expert surgeons have better operative skills than novices. Additionally, a chapter is dedicated on the evaluation of the mental stress using biomechanical information, based on the hypothesis that induced mental stress can affect the learning curve of trainees. The performance evaluation for the intubation is analyzed with different laryngoscopes and physicians, given the hypothesis that modern video-laryngoscopes help in improving the performance of the practice. The results are extremely promising, and show that the skill dexterity and competence could be effectively assessed by using the developed systems.

This thesis contains two primary parts: Part I introduces the development of the hardware and biomechanical model, together with the methodology used in the experiments, and it is composed by Chapters 1-4; Part II deals with the evaluation of medical operative skills based on analysis of subject’s upper body biomechanics in the fields of laparoscopic and airway trainings, and it is composed by the Chapters 5-7. Chapter 8 concludes.

The thesis is laid out as follows:

I. To develop a biomechanical model, combining the information of IMU and sEMG sensors, for evaluating the operative performance during medical training and clinical treatment:

• Chapter 1 introduces the research background with a detailed analysis of the state of the art of related researches on the assessment for medical operative skill evaluation by motion analysis and muscle force analysis. After that, the goals, methodology, innovation, and significance of this thesis are discussed.

• Chapter 2 is dedicated to the improvement of the WB-4 IMU developed in previous research, from hardware and software points of view. In particular, it is presented the new WB-4R system, based on hybrid wired-wireless approach with larger battery, able to sustain experiments up to five hours, a tremendous improvement with respect to the previous one hour of autonomy. A new online calibration of the magnetic sensor is then introduced, to make the sensor more robust to the presence of external ferromagnetic material interference, with the advantage of real time capability.

Finally, an algorithm for the joint angle estimation is presented, based on Extended Kalman Filter.

(4)

No.3

• Chapter 3 introduces the development of a new low-cost wireless sEMG sensor and the experimental results for its characterization. The commercial sEMG sensors are very expensive and they usually do not open the communication protocol, making it very difficult the synchronization with different sensors. The proposed design aims at overcoming both of the restrictions.

Additionally, this chapter tackles a problem that usually affects the sEMG signal: the baseline noise.

It is presented a new method to denoise real-time the sEMG signal using the same baseline to estimate the thresholds of the Wavelet denoising algorithm, named Baseline Adaptive Denoising Algorithm (BADA). The algorithm is verified during the peg-board training, used to improve the manual skill of surgeons for operative medical practice.

• Chapter 4 presents a new biomechanical EMG+IMU driven musculoskeletal model, developed to go beyond the two main limitations that restrict those models in real applications: the first is the space limitation, because usually optical tracking systems are used; the second is the absence of real-time capability. The wireless inertial measurement units and the surface EMG sensors described in the previous chapters help in overcoming the first limitation. A simplified original algorithm is described to achieve real time capability. The chapter closes with the experimental validation of the model for wrist flexion/extension.

II. To evaluate the medical operative skills during regular training by the proposed skill evaluation system:

• Chapter 5 describes the use of the biomechanical model to evaluate the expert and novice surgeons’

operative skills during regular laparoscopic training tasks. In collaboration with Kyushu University Medical Department, this is the first work in the world assessing a complex curriculum based on dry boxes, Virtual Reality and in-vivo surgical training. The results show significant differences not only between the novice and expert surgeons, but also between the training tasks.

• Chapter 6 presents the application of the proposed evaluation system in case of induced mental stress during laparoscopic training. Standard trainings are carried out to improve only the manual skill of the doctors, without any additional mental stress, that is actually present during the real operation. In this Chapter a novel protocol to induce mental stress is introduced and verified experimentally, showing finally that the mental stress had significant effects on the learning curve and number of errors.

• Chapter 7 presents the application of the biomechanical model to the endotracheal intubation medical practice, in which novice and expert physicians are compared together with different kinds of laryngoscopes. It is demonstrated that video-laryngoscopes reduce significantly the force applied during the insertion of the tube in both novices and experts.

Chapter 8 concludes the evaluation results and considerations in this thesis, and discusses about future works, which are proposed as a perspective to continue the research direction; in particular, a further miniaturization of the system and the enhancement of the real-time capability, could lead to real application also inside the operating and emergency room.

(5)

No.1

᪩

᪩✄⏣኱Ꮫ ༤ኈ㸦ᕤᏛ㸧 Ꮫ఩⏦ㄳ ◊✲ᴗ⦼᭩

(List of research achievements for application of doctorate (Dr. of Engineering), Waseda University) Ặ ྡ ( Luca BARTOLOMEO ) ༳

㸦As of July, 2013㸧

✀㢮ู 㢟ྡ Ⓨ⾲

ⓎⓎ⾜ᥖ㍕ㄅྡ

Ⓨ⾲

Ⓨ⾜ᖺ᭶

㐃ྡ⪅

㸦㸦⏦ㄳ⪅ྵࡴ㸧 1. ㄽᩥ

ۑ

Biomechanical Evaluation of the Phases during Simulated Endotracheal Intubation (ETI): Pilot Study on the Effect of Different Laryngoscopes

Proceedings of 2013 Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC, pp. 4887-4890

2013ᖺ07᭶ (published)

Luca Bartolomeo, Yohan Noh, Yusuke Kasuya, Mirei Nagai,

Massimiliano Zecca, Salvatore Sessa, Sarah Cosentino, Kohei Saito, Zhuohua Lin, Hiroyuki Ishii, Atsuo Takanishi Induced Mental Stress in

Peg Board Training:

Performance Evaluation by Motion and Muscle Analysis

Proceedings of 27^th

International Congress and Exhibition on Computer Assisted Radiology and Surgery CARS 2013 pp. S162-S163

Luca Bartolomeo, Massimiliano Zecca, Salvatore Sessa, Zhuohua Lin, Hiroyuki Ishii, Atsuo Takanishi

ۑ

Wavelet thresholding technique for sEMG denoising by baseline estimation

International Journal of Computer Aided Engineering and

Technology, Vol. 4, Num.

6, pp. 517-534

Luca Bartolomeo, Massimiliano Zecca, Salvatore Sessa, Atsuo Takanishi

A Methodology for the Performance Evaluation of Inertial Measurement Units

Journal of Intelligent &

Robotic Systems, pp. 1-15 2012ᖺ09᭶ (published)

Salvatore Sessa, Massimiliano Zecca, Zhuohua Lin, Luca Bartolomeo, Hiroyuki Ishii, Atsuo Takanishi Online magnetic calibration

of a cutting edge 9-axis wireless Inertial Measurement Unit

International Journal of Applied Electromagnetics and Mechanics, Vol. 39, Num. 1, pp.779-785

Luca Bartolomeo, Massimiliano Zecca, Salvatore Sessa, Zhuohua Lin, Hiroyuki Ishii, Atsuo Takanishi

ۑ

Biomechanical analysis of induced mental stress in laparoscopy surgical training by surface Electromyography

Proceedings of 2012 4th IEEE RAS & EMBS International Conference on Biomedical Robotics and Biomechatronics (BioRob), pp. 1194-1198

Luca Bartolomeo, Massimiliano Zecca, Salvatore Sessa, Zhuohua Lin, Hiroyuki Ishii, Hao Xu,

Munemori Uemura, Yoshihiro Nagao Morimasa Tomikawa, Makoto Hashizume Atsuo Takanishi

(6)

No.2

᪩

᪩✄⏣኱Ꮫ ༤ኈ㸦ᕤᏛ㸧 Ꮫ఩⏦ㄳ ◊✲ᴗ⦼᭩

(List of research achievements for application of doctorate (Dr. of Engineering), Waseda University)

Ⓨ⾲

Ⓨ⾜ᖺ᭶

㐃ྡ⪅

㸦㸦⏦ㄳ⪅ྵࡴ㸧 ㄽᩥࡢ⥆ࡁ

Baseline Adaptive Wavelet Thresholding Technique for sEMG Denoising

Proceedings of 2011 International Symposium on Computational Model for Life Science, pp.

205-214

Luca Bartolomeo, Massimiliano Zecca, Salvatore Sessa, Zhuohua Lin, Hiroyuki Ishii, Yoshikazu Mukaeda, Atsuo Takanishi Development of the

Wireless Ultra-miniaturized Inertial Measurement Unit WB-4: Preliminary Performance Evaluation

Proceedings of 2011 Annual International Conference of the IEEE Engineering in Medicine and Biology

Society,EMBC, pp. 6927 – 6930

Zhuohua Lin, Massimiliano Zecca, Salvatore Sessa, Luca Bartolomeo, Hiroyuki Ishii, Atsuo Takanishi

Surface EMG and Heartbeat Analysis Preliminary Results in Surgical Training: Dry Boxes and Live Tissue

Proceedings of 2011 Annual International Conference of the IEEE Engineering in Medicine and Biology

Society,EMBC, pp.

1113-1116

Luca Bartolomeo, Zhuohua Lin, Massimiliano Zecca, Salvatore Sessa, Hiroyuki Ishii, Hao Xu,

Munemori Uemura, Morimasa Tomikawa, Makoto Hashizume Atsuo Takanishi Waseda Bioinstrumentation

system WB-3 as a wearable tool for objective

laparoscopic skill evaluation

Proceedings of 2011 IEEE International Conference on Robotics and

Automation (ICRA), pp.

5737 - 5742

Zhuohua Lin, Munemori Uemura, Massimiliano Zecca, Salvatore Sessa, Hiroyuki Ishii, Luca Bartolomeo, Kazuko Itoh,

Morimasa Tomikawa, Makoto Hashizume Atsuo Takanishi Ultra-miniaturized WB-3

Inertial Measurement Unit:

performance evaluation of the attitude estimation

Proceedings of 2010 IEEE International Conference on Robotics and

Biomimetics, pp. 998-1003, 2010.

Salvatore Sessa, Massimiliano Zecca, Zhuohua Lin, Luca Bartolomeo, Kazuko Itoh, Hiroyuki Ishii, Yoshikazu Mukaeda, Yuto Suzuki,

Atsuo Takanishi

(7)

No.3

᪩

᪩✄⏣኱Ꮫ ༤ኈ㸦ᕤᏛ㸧 Ꮫ఩⏦ㄳ ◊✲ᴗ⦼᭩

(List of research achievements for application of doctorate (Dr. of Engineering), Waseda University)

Ⓨ⾲

Ⓨ⾜ᖺ᭶

㐃ྡ⪅

㸦㸦⏦ㄳ⪅ྵࡴ㸧 2. ㅮ₇

On the development of a new protocol to induce stress during Surgical Training: EMG analysis

2012 Annual International Conference of the IEEE Engineering in Medicine and Biology

Society,EMBC

2012ᖺ08᭶ Luca Bartolomeo, Massimiliano Zecca, Salvatore Sessa, Zhuohua Lin, Hiroyuki Ishii,

Morimasa Tomikawa, Makoto Hashizume Atsuo Takanishi Development of a Cutting

Edge 9-axis Wireless Inertial Measurement Unit

ISEM 2011 - 15th International Symposium on Applied

Electromagnetics and Mechanics

2011ᖺ09᭶ Luca Bartolomeo, Massimiliano Zecca, Salvatore Sessa, Zhuohua Lin, Hiroyuki Ishii, Atsuo Takanishi

ே㛫ࡢ㐠ື࠾ࡼࡧ⏕⌮ᣦ ᶆࡢィ ࢩࢫࢸ࣒࡟㛵ࡍ

ࡿ◊✲

➨29ᅇ᪥ᮏࣟ࣎ࢵࢺ Ꮫ఍

2011ᖺ09᭶ ᩪ⸨⯟ᖹ, 㕥ᮌᝆே,

ࣝ࢝ ࣂࣝࢺ࣓ࣟ࢜, ᯘࢯ࢘࣡,

ఀ⸨ຍᑑᏊ,

▼஭⿱அ,

ࢭࢵࢧ ࢧࣝࣂࢺ࣮ࣞ,

᳜ᮧ᐀๎,

ᐩᕝ┒㞞, ᶫ∎ㄔ,

࣐ࢵࢩ࣑ࣜ࢔࣮ࣀ

ࢮࢵ࢝, 㧗す῟ኵ, New Surface

Electromyography Denoising Technique with Baseline Noise Estimation

4th International Workshop

on Cybernics (IWC2011) 2011ᖺ03᭶ Luca Bartolomeo, Massimiliano Zecca, Atsuo Takanishi

3. ㈹

T. J. Tarn Best Paper in

Robotics Finalist 2010 IEEE International Conference on Robotics and Biomimetics (ROBIO2010)

2010ᖺ12᭶ Salvatore Sessa, Massimiliano Zecca, Zhuohua Lin, Luca Bartolomeo, Kazuko Itoh, Hiroyuki Ishii, Yoshikazu Mukaeda, Yuto Suzuki,

Atsuo Takanishi

Graduate School of Advanced Science and Engineering, Waseda University