用于脑瘫儿下肢康复装置的表面肌电信号特征提取方法研究

摘要

采用时域、频域、时频域和熵的特征提取方法,找到适合脑瘫儿表面肌电信号的特征提取方法.通过在训练过程加一个阻力得到四个不同肌肉活性的训练阶段数据,对数据进行预处理和特征提取,然后用因子分析法对所提取的特征进行分析,实验结果显示本研究所提出的时域、时频域和熵特征部分适用于脑瘫儿,频域特征不适用于脑瘫儿.本研究结果对脑瘫儿的康复训练有很大的帮助.%To obtain the representative features using time domain, frequency domain, time-frequency domain and nonlinearity analysis feature extraction methods.A force disturbance was applied to the right leg during swing phase of gait cycle while a subject walked on a treadmill.Surface electromyography signals(sEMG) from 8 leg muscles were recorded in three loading conditions, i.e., baseline, with load, and load release.The factor analysis results showed that part of time domain, entropy and wavelet packet features were suitable for assessing sEMG signal features in children with cerebral palsy(CP).The results may provide contributions to establish an EMG based rehabilitation system for children with CP.