基于机器学习的下肢被动康复训练过程监测评价研究  

Monitoring and evaluation machine⁃learning⁃based passive rehabilitation training process of lower limbs

作　　者：吴达 马超[1] 高经纬 谷玉海[1] WU Da;MA Chao;GAO Jingwei;GU Yuhai(MOE Key Laboratory of Modern Measurement and Control Technology,Beijing Information Science and Technology University,Beijing 100192,China)

机构地区：[1]北京信息科技大学现代测控技术教育部重点实验室,北京100192

出　　处：《现代电子技术》2022年第16期175-180,共6页Modern Electronics Technique

基　　金：北京市属高校高水平创新团队建设计划项目(IDHT20180513)。

摘　　要：针对人体下肢的被动康复训练过程,文中使用某下肢康复训练机器人进行下肢被动训练,训练过程中采集下肢股直肌和腓肠肌的sEMG信号和大腿的运动姿态信号,运用机器学习方法对运动姿态信号和表面肌电(sEMG)信号进行分析,实现了下肢中4种不同被动训练状态的识别及其对下肢肌群训练康复效果的评价。结果显示:联合使用IMU和sEMG进行下肢被动训练过程监测,并通过机器学习算法进行处理,可以实现不同监测过程的自动识别以及下肢肌群训练效果量化分析判断。研究结果可为实现基于下肢被动康复训练过程的智能控制与康复情况评价奠定研究基础。For the process of human lower limbs passive rehabilitation training,a lower limbs rehabilitation training robot is used in the study to conduct lower limbs passive training.The SEMG signals of rectus femoris and gastrocnemius muscles of lower limbs and the motion posture signals of thigh are collected during the training,and the machine learning method is used to analysis these signals,so as to realize the recognition of four different passive training states in the lower limb and the evaluation of the training rehabilitation effect on lower limbs muscle group.The results show that the IMU and sEMG are combined to monitor the process of lower limbs passive training,and the machine learning algorithm is used to process,which can realize the automatic recognition of different monitoring process,and the quantitative analysis and judgment of lower limbs muscle group training effect.The results can lay a research foundation for the realization of intelligent control and rehabilitation evaluation based on the passive rehabilitation training of lower limbs.

关 键 词：下肢被动训练 机器学习 IMU SEMG 康复训练评价 信号处理 智能控制 康复训练机器人 

分 类 号：TN919-34[电子电信—通信与信息系统] TP311[电子电信—信息与通信工程]