Electromyogram signal analysis using eigenvalue decomposition of the Hankel matrix

Abstract

The identification of neuromuscular abnormalities can be performed using electromyogram (EMG) signals. In this paper, we have presented a method for the analysis of amyotrophic lateral sclerosis (ALS) and normal EMG signals. The motor unit action potentials (MUAPs) have been extracted from EMG signals. The proposed method is based on improved eigenvalue decomposition of the Hankel matrix (IEVDHM). Two significant decomposed components obtained from IEVDHM, are considered for analysis purpose. These components are obtained on the basis of higher energy of components. Correntropy (CORR) and cross-information potential (CIP) are computed for two components. Thereafter, statistical analysis has been performed using the Kruskal–Wallis statistical test. We have observed that the IEVDHM method is able to provide the components, which can distinguish the ALS and normal EMG signals using CORR and CIP parameters. © Springer Nature Singapore Pte Ltd 2019.