Robust muscle activity onset detection using an unsupervised electromyogram learning framework.

Jie Liu,Ping Zhou,Dongwen Ying,William Z. Rymer,Mikhail A. Lebedev

doi:10.1371/journal.pone.0127990

Abstract

Accurate muscle activity onset detection is an essential prerequisite for many applications of surface electromyogram (EMG). This study presents an unsupervised EMG learning framework based on a sequential Gaussian mixture model (GMM) to detect muscle activity onsets. The distribution of the logarithmic power of EMG signal was characterized by a two-component GMM in each frequency band, in which the two components respectively correspond to the posterior distribution of EMG burst and non-burst logarithmic powers. The parameter set of the GMM was sequentially estimated based on maximum likelihood, subject to constraints derived from the relationship between EMG burst and non-burst distributions. An optimal threshold for EMG burst/non-burst classification was determined using the GMM at each frequency band, and the final decision was obtained by a voting procedure. The proposed novel framework was applied to simulated and experimental surface EMG signals for muscle activity onset detection. Compared with conventional approaches, it demonstrated robust performance for low and changing signal to noise ratios in a dynamic environment. The framework is applicable for real-time implementation, and does not require the assumption of non EMG burst in the initial stage. Such features facilitate its practical application.

Highlights

The function of muscle activity onset detector using surface electromyogram (EMG) is to distinguish occurrence of active muscle activity from baseline
The muscle activity onset detection method developed in this study is to model the subband logarithmic energy of EMG signal using an unsupervised learning framework based on Gaussian mixture model (GMM)
This study develops a statistical framework based on unsupervised learning to model the EMG burst and non-burst distributions in the frequency domain for muscle activity onset detection

Summary

Introduction

The function of muscle activity onset detector using surface electromyogram (EMG) is to distinguish occurrence of active muscle activity from baseline. The unsupervised learning framework was developed based on a sequential Gaussian mixture model (GMM) It utilizes the energy distribution in Mel-spaced frequency bands of the signal as its feature parameter. The unsupervised learning based on a sequential GMM was applied to both simulated and experimental surface EMG signals to evaluate its performance for muscle activity onset detection. The muscle activity onset detection method developed in this study is to model the subband logarithmic energy of EMG signal using an unsupervised learning framework based on GMM. To evaluate the performance of the unsupervised learning framework for detection of muscle activation, a series of surface EMG signals were simulated by filtering white Gaussian noise with a shaping filter modeling the characteristics of typical surface EMG [27]. All the simulated signals were sampled at 2 kHz and processed with a 6th order Butterworth band-pass filter at 20–500 Hz

Different methods used for performance comparison

Discussion