Publication: Game-based Enhancement for Rehabilitation Based on Action Recognition Using Kinect
Issued Date
2019-02-22
Resource Type
ISSN
21593450
21593442
21593442
Other identifier(s)
2-s2.0-85063195234
Rights
Mahidol University
Rights Holder(s)
SCOPUS
Bibliographic Citation
IEEE Region 10 Annual International Conference, Proceedings/TENCON. Vol.2018-October, (2019), 303-308
Suggested Citation
Chanat Sinpithakkul, Worapan Kusakunniran, Sunee Bovonsunthonchai, Peemongkon Wattananon Game-based Enhancement for Rehabilitation Based on Action Recognition Using Kinect. IEEE Region 10 Annual International Conference, Proceedings/TENCON. Vol.2018-October, (2019), 303-308. doi:10.1109/TENCON.2018.8650226 Retrieved from: https://repository.li.mahidol.ac.th/handle/123456789/50649
Research Projects
Organizational Units
Authors
Journal Issue
Thesis
Title
Game-based Enhancement for Rehabilitation Based on Action Recognition Using Kinect
Other Contributor(s)
Abstract
© 2018 IEEE. The physical rehabilitation is a way to improve physical abilities of patients. One of the main problems is patients can get bored from repeated rehab-activities which must be worked out for a long period or even for a lifetime. This can decrease patients' motivations to do the rehabilitation. In addition, game is considered as an entertainment media that is able to use to motivate players to engage in particular activities by giving rewards in exchange. Therefore, this paper proposes the digital game-based enhancement for the Cerebral Palsy (CP) rehabilitation. The CP is the congenital disease caused by the damage on a part of the brain that controls the movement functionality. In this study, the patient must perform specific rehab-actions to control the games. Four games are developed for four rehab-actions including shoulder flexion, shoulder abduction, shoulder horizontal abduction and elbow flexion/extension. They are parts of fundamental movements needed for the Activities of Daily Living (ADL). In each game, the patient must perform the requested rehab-action naturally without carrying any controller. The Kinect is used to track the movements from the patient. Then, the rule-based system is trained and used to detect the correct rehab-action for controlling the game. Then, the experiment is performed to see possibility of using the game-based system for the rehabilitation enhancement. It shows the promising performance in terms of action recognition accuracy that is flexible enough for the patient, the play records, and the expert review.