基于肌电和惯性传感器数据融合的脑卒中患者上肢够物运动定量评估

关键词：微传感器人体运动捕获  肌电  上肢康复

Keywords:micro-sensor motion capture  EMG  rehabilitation of upper limb

摘要：

      摘要 目的:探讨融合肌电和惯性传感数据的定量化脑卒中患者上肢运动评估方法的有效性。 方法:实验联合使用微型惯性传感器和多通道肌电传感器，同步采集了20例脑卒中患者（患者组）和年龄匹配的10名老年人（健康组）的肌电及运动数据，融合这两种数据分别进行了分组统计学分析和典型个例分析。 结果:患者组的运动学参数与健康组比较差异有显著性意义，并且运动学参数与肌电参数之间有着较高的相关性。运动参数所反映出的异常运动模式与肌电所反映出的情况相一致，肌电数据能用于解释异常运动模式的原因。 结论:本定量化康复评估方法可以实时、定量和可视化地进行临床分析和诊断治疗。

Abstract:

      Abstract Objective: To explore the effectiveness of rehabilitation assessing method in stroke patients during upper limb reaching based on the fusion of EMG and inertial sensing data. Method: In experiment inertial micro-sensor and multi-channel EMG sensor were combined together to synchronously collect the EMG and motion data from 20 stroke patients(patients group) and 10 age-matched old people(healthy group). Result: Packet statistical analysis and typical case analysis were done by fusing EMG and motor data. The difference of kinematic parameters between patients group and healthy group had statistical significance, and kinematic parameters had high correlation with EMG parameters. Abnormal movement patterns reflected by motion parameters were consistent with the circumstances reflected by EMG parameters. And the EMG parameters could explain the cause why the movement patterns were abnormal. Typical case analysis could accurately reflect all joint angles and the coordination between six muscles at any time during upper limb reaching process. Conclusion: The experiment proved the effectiveness and value of this quantitative rehabilitation assessment method. This method can perform the real-time, quantitative and visualized clinical analysis and diagnosis which is of great significance in clinical applications.