Frequency Feature for Proportional Myoelectric Control of Robotic Rehabilitation Therapy
Jong-Hoon Kim✽, Hae Yean Park✽✽, Hye-Min Kang✽✽✽, YoungJin Jung✽✽✽✽ ✽Dept of Computer Science, Kent State University, Kent, OH, USA ✽✽Dept. of Occupational Therapy, College of Health Science, Yonsei University, South Korea ✽✽✽Linguistics, University of Minnesota, Twin Cities, MN, USA ✽✽✽✽Dept. of Radiological Science at Health Science Division, Dongseo University, Busan, South Korea
국문초록 Objective :There have been active movements in developing exoskeletal robot system (ERS) which is electrically powered, especially for after stroke patient’s rehabilitation purposes. For the general principle of motor skill rehabilitation, controlling algorithm for ERS should simulate actual tasks based on surface electromyogram (sEMG). Methods :In this study, the modified linear regression model based on frequency feature was demonstrated, and compared to conventional method. An electro-goniometer was used to measure a degree of wrist movement, and six channel sEMG device were employed to measure electric activates on the middle of forearm. Results :The results show that modified linear regression model has higher accuracy for controlling ERS. According to the results, we assure of the possibility that the suggested model will be used for stroke robot rehabilitation. Conclusion :Moreover, these techniques will be used as a prominent tool in making device or finding new therapy approaches in robot assisted rehabilitation for stroke survivors.