远程康复中穿戴式传感器的运动捕获算法设计及应用  

Design and Application of Motion Capturing Algorithm for Wearable Sensors in Telerehabilitation

作　　者：董梁 屈云[2,3,4] DONG Liang;QU Yun(School of Data Science and Engineering,South China Normal University,Shanwei Guangdong 516600,China;Department of Rehabilitation Medicine,West China Hospital,Sichuan University,Chengdu Sichuan 610041,China;School of Rehabilitation Sciences,West China School of Medicine,Sichuan University,Chengdu Sichuan 610041,China;Sichuan Provincial Key Laboratory of Rehabilitation Medicine,Sichuan University,Chengdu Sichuan 610041,China)

机构地区：[1]华南师范大学数据科学与工程学院,广东汕尾516600 [2]四川大学华西医院康复医学科,四川成都610041 [3]四川大学华西康复医学院,四川成都610041 [4]四川省康复医学重点实验室,四川成都610041

出　　处：《中国医疗设备》2021年第11期32-35,39,共5页China Medical Devices

基　　金：国家重点研发计划(2017YFC1308504)。

摘　　要：本文提出了一种由穿戴式MEMS传感器和移动终端组成的自由运动捕获设备,康复医生可使用这一设备以远程方式查看康复患者的日常运动状况,进行运动功能评估并做出定制化的康复训练指导。设备根据不同肢体的运动特性设计了不同的卡尔曼滤波算法,将穿戴式传感器的量测值转化为三维空间的姿态角,在移动终端里将各个肢体的姿态角融合后以3D模型方式实时显示患者的运动状况。此设备部署在华西医院院外康复患者家中进行测试,初期实验结果证实此算法相对于光学捕获方法的平均角度差异为4.0°,平均延迟为297 ms。移动终端所显示的模型动作可实时、准确地呈现患者的运动姿态。这一设备和算法为离院康复患者在专业医生监督下进行康复训练提供了可能,为康复医生展开远程康复医疗提供了新的工具和思路。An ambulatory movement capture device is proposed in this paper,which is comprised of wearable MEMS sensors and mobile terminal.The device enables doctors to monitor daily activities of rehabilitation patients,evaluate their motor functions and prescribe rehabilitation training program in remote manner.With consideration of the motor functions of different body segments,multiple Kalman filtering algorithms are designed to convert the sensor measurements to 3D rotation angles.After the rotation angles of each limb were fused in mobile terminal,the motion status of the patient was displayed in real-time in the form of 3D model.The device was deployed in the home of out-of-hospital rehabilitation patients of West China Hospital.Preliminary experimental results indicate that the average difference of tracing angle was 4.0°between the proposed algorithm and optical capture approach,and the average delay of the system was 297 ms.The model activities presented on the mobile terminal indicate that the motion of the patient can be accurately captured on real-time basis.As a conclusion,the device and the proposed algorithm make it possible that out-of-hospital rehabilitation patients were able to conduct rehabilitation training under the supervision of professionals.It provides new tools and ideas for doctors to carry out telerehabilitation medicine.

关 键 词：穿戴式传感器 卡尔曼滤波 运动捕获 康复医学 远程康复 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]