Pre-processing and classification of EMG signals by using modern method

motor unit action potentials (MUAPs) from intramuscular electromyographic signals. The proposed method automatically detects the number of template MUAP clusters and classifies them into normal, neuropathic or myopathic. To extract a feature vector from the EMG signal, we use different AR parametric methods and features of signals. The approach has been validated using a dataset of EMG recordings composed of 1200 MUAPs obtained from 7 normal subjects, 7 subjects suffering from myopathy, and 13 subjects suffering from neurogenic disease. The correct identification rate for MUAP clustering is 97, 90 and 87% for normal, myopathic and neuropathic, respectively. Almost ninety percent of the superimposed MUAPs were correctly identified. The obtained accuracy for MUAP classification is about 92% for combined neural network. The proposed method, apart from efficient EMG decomposition addresses automatic MUAP classification to neuropathic, myopathic or normal classes directly from raw EMG signals. A similar classification was also made with FEBANN in the study. Obtained results show that the accuracy rates for CNN in this study is higher than FEBANN