Extracting time-frequency feature of single-channel vastus medialis EMG signals for knee exercise pattern recognition.

Yi Zhang,Steven W Su,Dezhong Yao,Peiyang Li,Peng Xu,Qing Guo,Xuyang Zhu,Yih-Kuen Jan

doi:10.1371/journal.pone.0180526

Yi Zhang, Steven W Su + Show 6 more

Open Access

https://doi.org/10.1371/journal.pone.0180526

Copy DOI

Abstract

The EMG signal indicates the electrophysiological response to daily living of activities, particularly to lower-limb knee exercises. Literature reports have shown numerous benefits of the Wavelet analysis in EMG feature extraction for pattern recognition. However, its application to typical knee exercises when using only a single EMG channel is limited. In this study, three types of knee exercises, i.e., flexion of the leg up (standing), hip extension from a sitting position (sitting) and gait (walking) are investigated from 14 healthy untrained subjects, while EMG signals from the muscle group of vastus medialis and the goniometer on the knee joint of the detected leg are synchronously monitored and recorded. Four types of lower-limb motions including standing, sitting, stance phase of walking, and swing phase of walking, are segmented. The Wavelet Transform (WT) based Singular Value Decomposition (SVD) approach is proposed for the classification of four lower-limb motions using a single-channel EMG signal from the muscle group of vastus medialis. Based on lower-limb motions from all subjects, the combination of five-level wavelet decomposition and SVD is used to comprise the feature vector. The Support Vector Machine (SVM) is then configured to build a multiple-subject classifier for which the subject independent accuracy will be given across all subjects for the classification of four types of lower-limb motions. In order to effectively indicate the classification performance, EMG features from time-domain (e.g., Mean Absolute Value (MAV), Root-Mean-Square (RMS), integrated EMG (iEMG), Zero Crossing (ZC)) and frequency-domain (e.g., Mean Frequency (MNF) and Median Frequency (MDF)) are also used to classify lower-limb motions. The five-fold cross validation is performed and it repeats fifty times in order to acquire the robust subject independent accuracy. Results show that the proposed WT-based SVD approach has the classification accuracy of 91.85%±0.88% which outperforms other feature models.

Highlights

The lower-limb motion is of vital importance in human daily living activities
Based on the segmented EMG data for all trials, the statistical analysis for these four time-domain features is evaluated by one-way analysis of variance analysis, which shows that Mean Absolute Value (MAV), RMS, iEMGP, and ZC have means significantly different for lower-limb motions (p < 0.05)
In order to evaluate the classification performance, there are five different feature vectors established for training and testing the lower-limb motions, including time-domain features (MAV +RMS+integrated EMG (iEMG)+ZC), frequency-domain feature (MNF+Median Frequency (MDF)), combined time and frequency features (MAV+RMS+iEMG+ZC+MDF+Mean Frequency (MNF)), Wavelet Transform (WT)-based Singular Value Decomposition (SVD) features, and combined time, frequency, and WT-based SVD features (MAV+RMS +iEMG+ZC+MDF+MNF+cD1+cD2+cD3+cD4+cD5+cA5)

Summary

Introduction

The lower-limb motion is of vital importance in human daily living activities. Power-assisted robotic systems have been developed to target these problems and assist people who need help in terms of their daily living [1,2,3,4,5,6]. Most of these systems are originally activated by EMG signals as they can directly indicate the electrophysiological responses to daily living activities. The electric indications of the activated motor units for one muscle contraction should be unique, and only vary due to brain activity. A single-channel-based EMG pattern classification is developed in order to provide an easy-to-use condition for detections of lower-limb motions

Objectives

Methods

Results

Conclusion