Classification and Diagnosis of Myopathy EMG Signals Using the Continuous Wavelet Transform

Electromyography (EMG) is a very suitable technique for the diagnosis of the skeletal muscle state. This work introduces a powerful tool for the analysis and the classification of myopathy and normal EMG signals. Continuous wavelet transform (CWT) was applied to a total of 406 EMG records. For the classification purpose, four statistical features were extracted from the continuous wavelet analysis. The classification is performed using k Nearest Neighborhood (k-NN) and Support Vector Machine (SVM) classifiers. 10-fold cross-validation method was used in this study since it delivered higher results. To evaluate the performances of the classifiers, metrics such as specificity, sensitivity and accuracy were calculated. Experimental results showed that both SVM and k-NN classifiers yielded high results. An accuracy value of 91.11 ±0.38 (Mean ± Standard Deviation) was obtained with k-NN classifier using 9 nearest neighbors. In regards to SVM, the classifier achieved an accuracy of 91.01±0.28 using the Radial Basis Function (RBF) kernel. The proposed technique demonstrated that it is an efficient tool in the discrimination between myopathic patients and normal subjects.