Wearable rehabilitation assessment system based on complex network

Abstract

For stroke patients, rehabilitation assessment performs an important reference for diagnosis and treatment in the rehabilitation process. In order to conduct the rehabilitation assessment quickly, accurately and objectively, a wearable multisource upper limb rehabilitation quantitative assessment system based on 9-axis sensors and flex sensors is designed. The data collected by different sensors were assessed quantitatively and classified by complex network algorithm. To verify the performance of the system, an experiment was carried out on four volunteers, including one healthy person and three stroke patients, whose clinical rehabilitation assessment were stage II, stage III and stage IV respectively. The complex network diagrams and metrics results of the volunteers' 10 Bobath handshake actions completed in unconstrained state were researched and analyzed. The results indicated that 10 Bobath handshake actions acted by stroke patients in different stage and healthy person showed significant difference in network connection quantity, average degree, average path length and average clustering coefficient.