Clustering analysis and pattern discrimination of EMG linear envelopes

A technique has been developed for performing pattern analysis of electromyographic (EMG) activities generated during locomotion. It was found that the shapes of the EMG linear envelopes (LE) are mainly determined by their phase spectra; their magnitude spectra are much less important. Autoregressive (AR) parametric models and discrete Fourier transform (DFT) approaches were tested and compared. The latter proved to be a better way to describe the EMG LEs. Feature extraction and clustering were performed by performing a DFT of EMG LEs, extracting part of the phase and magnitude spectra as features, and using the percent powers to weigh the corresponding harmonics. The approach was applied to the clustering analysis of EMG LEs of normal and anterior cruciate ligaments (ACL) injured subjects during walking.