Myopathy Detection and Classification Based on the Continuous Wavelet Transform

Published online: Nov 25, 2019
Full Text: PDF (1.12 MiB)
DOI: 10.24138/jcomss.v15i4.796
Authors:
Abdelali Belkhou, Abdelouahed Achmamad, Atman Jbari

Abstract

Electromyography (EMG) is the study of the electrical activity of the muscle. This technique is often used in the diagnosis of neuromuscular diseases. Myopathy is one of these cases, which affect the muscle and causes many changes in the electromyography signal characteristics. This paper presents a new method for analysis and classification of normal and myopathy EMG signals based on continuous wavelet transform (CWT). Classification algorithms, namely Support Vector Machine (SVM), k-Nearest Neighbor (k-NN), Decision Tree (DT), Discriminant Analysis (DA) and Naïve Bayes (NB) were used as classifiers in our study. Five Features were extracted from the continuous wavelet analysis and used as inputs to the mentioned classifiers. Comparison between different classification methods developed in this study was made by evaluation of their results based on multiple scalar performances, mainly accuracy, sensitivity, and specificity. Different combinations of features with different kernel functions were discussed to achieve better performances. Results showed that k-NN classifier achieved the best performances with an accuracy value of 93.68%.

Keywords

Electromyography (EMG), continuous wavelet transform (CWT), support vector machine (SVM), k-nearest neighbor (k-NN), decision tree (DT), discriminant analysis (DA), native bayesian (NB)
Creative Commons License 4.0
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.