Neuromuscular disease classification based on discrete wavelet transform of dominant motor unit action potential of EMG signal

Electromyogram (EMG) is a recording of the electrical activity of skeletal muscles. These signals are used in the medical field for diagnosis of various diseases. In this paper neuromuscular diseases are classified by using discrete wavelet transform. Here dominant motor unit action potentials are extracted from the EMG signals via template matching based decomposition method. Apart from considering all motor unit action potentials, dominant motor unit action potential is considered for disease classification. Because all MUAPs are not uniquely represents a class. Therefore, dominant MUAP based on an energy criterion is proposed for feature extraction. Then statistical features are extracted from dominant MUAP by decomposing it to produce wavelet coefficients. Finally K-nearest neighbor classifier (KNN) is used to classify neuromuscular diseases.