Wearable Motion Recognition System Based on Dynamic Time Warping

6th Annual 2018 International Conference on Geo-Spatial Knowledge and Intelligence

Guo, Liquan^1 ; Chen, Jing^1 ; Zheng, Qi^1 ; Wang, Jiping^1 ; Bian, Jieyong^2 ; Wang, Xiaojun^2 ; Ouyang, Ruizhi^2

Suzhou Institute of Biomedical Engineering and Technology, Chinese Academy of Sciences, Suzhou, China^1

Suzhou Xiangcheng People's Hospital, Suzhou, China^2

关键词: Automated systems;    Dynamic time warping algorithms;    Motion recognition;    Motion segmentation;    Motion segments;    Recognition accuracy;    Rehabilitation training;    Stroke patients;   

Others  :  https://iopscience.iop.org/article/10.1088/1755-1315/234/1/012087/pdf
DOI  :  10.1088/1755-1315/234/1/012087

来源: IOP

【 摘 要 】

In order to monitor the rehabilitation training of stroke patients in unsupervised situation and provide rehabilitation advice for rehabilitation clinicians, a wearable wireless motion recognition system has been developed using 9-axis wearable sensors, to identify patients' typical upper limb movements, e.g. Bobath handshake, stretch elbow and press hand, shoulder joint horizontal outreach, elbow buckling contact, paraplegia hand touch the shoulder, sequine pressure rotary before supination. After the original data is collected and preprocessed, each rehabilitation training action is segmented. Dynamic Time Warping (DTW) algorithm is used to calculate the similarity between each motion segment and the standard template data, and the motion recognition is carried out according to the calculated results. To verify the performance of the system, 20 stroke patients were recruited as volunteers. Each patient wore a 9-axis wearable sensor on his upper limb and performed six rehabilitation training exercises. After motion segmentation, 1400 motion fragments were obtained and used as samples to test the system. It has been found that the recognition accuracy of the system for the six rehabilitation training exercises is over 90%. This result provides a well reference for further development of an automated system for stroke patient rehabilitation motion recognition.