EMG BASED NEUROMUSCULAR DISEASE CLASSIFICATION

Abstract

Electromyography (EMG) is the recording of electrical activity of the skeletal muscles. Various muscle related and neuromuscular diseases are diagnosed by analyzing an EMG signals. Myopathy, Neuropathy and Normal subjects are analyzed by using discrete wavelet transform (DWT). In time domain analysis, root mean square value is calculated to classify neuromuscular diseases. But DWT based feature extraction scheme gives the better results than RMS value because in that case, signal analysis is carried out both in time and frequency domain simultaneously. EMG signal is divided into a number of frames and a signal analysis is performed in frame by frame manner. DWT based feature extraction scheme is utilized for feature extraction so as to separate normal person to that of diseased patients. Higher valued DWT coefficients are considered by arranging these coefficients in descending order which are used for feature extraction. Maximum and average value of first five higher valued coefficients is calculated to reduce feature dimension. But even for better classification of these two main neuromuscular diseases namely Myopathy and Neuropathy and Healthy signals is performed using cross-correlation based feature extraction technique. For this purpose, cross-correlation of Healthy, Myopathy and Neuropathy disease EMG signal is done with a reference Healthy signal. Selective features like Hjorth, and statistical features comprising mean, standard deviation and power are extracted from the cross-correlated signals. Support Vector Machine (SVM) and k-Nearest Neighbor (kNN) are the two classifiers used for this work. Higher classification accuracy is obtained using SVM.