The Centre for Pattern Analysis and Machine Intelligence in association with IEEE Computational Intelligence and Systems, Man & Cybernetics Society Chapters of KW section, invite you to attend the following seminar:
Toward the Minimum Measured-input Models for the Evaluation of Human Motor Ability
Dr. Vincent Bonnet, Japanese Society for Promotion of Research, Tokyo University of Agriculture and Technology, Tokyo, Japan
University of Waterloo
William G. Davis Computer Research Centre
Room DC-1302
Refreshments will be provided
Abstract:
The problematic of evaluating and understanding the physical functional limitations of patients suffering from postural pathologies and establishing the relationship between individual disability and biomechanics will be addressed in this talk.
Starting from simple experimental paradigms used in rehabilitation field predictive models, based on biomechanical analysis, optimization processes and classical robotics tools, have been developed. These models allow, at different levels of description in actuator dynamics and sensory feedback, a better understanding of the optimal behaviors and constraints acting on the postural system during rehabilitation exercises. After experimental validations on humans and humanoid robots, these models have been extended to capture specific invariants observed with hemiplegic patients.
In order to use the outputs of these models in clinical practice and daily life practice, new and affordable measurement methodologies for human movement have been developed in the purpose of minimizing the complexity of experimental setup. In these approaches, using a single inertial sensor and a RGB-depth camera, joint variables and segmental mechanics could be accurately estimated opening the way of future in-home rehabilitation applications.
Finally, a new friendly-user identification method of subject specific inertial parameters using affordable sensors will be presented.
These computational models together with the corresponding affordable and friendly-user measurement methodologies could spread among clinicians. This strong potential for future rehabilitation applications could contribute to a general better understanding of the relationships between biomechanics and disability.
Biography:
Vincent Bonnet received the B.E. degree in electrical engineering, in 2005, and the Ph.D. degree in automatics control and robotics, in 2009, both from the Science University of Montpellier 2. From 2010 to 2012, he has been working as a post-doc in Rome, Italy under the direction of Professor Aurelio Cappozzo, historical reference in Biomechanics and motion analysis. In 2013 he was teaching assistant at the University of Montpellier and belonged to the Euromov Institute where he was in charge of the research related to motion analysis.
He is currently a post-doc for the Japanese Society for Promotion of Research at the Tokyo University of Agriculture and Technology, Tokyo, Japan. His research interests include various aspects of low-cost sensors, motion analysis, humanoid robotics, system identification and control. He teaches courses in kinematics and dynamics, automatic and biomechanics.
All are welcome to attend!