Surface EMG signal segmentation based on HMM modelling: Application on Parkinson’s disease

The study of burst electromyographic (EMG) activity periods during muscles contraction and relaxation is an important challenging problem. It can find several applications like movement patterns analysis, human locomotion analysis neuromuscular pathologies diagnosis such as Parkinson disease. This paper proposes a new frame work for detecting the onset (start) / offset (end) EMG by segmenting signal in regions muscle (AC) non (NAC) using Discrete Wavelet Transform (DWT) feature extraction Hidden Markov Models (HMM) classification AC NAC classes. objective this to design efficient segmentation system signals recorded from Parkinsonian group control (healthy). results evaluated on ECOTECH project database principally Accuracy (Acc) error rate (Re) criterion show highest performance HMM models 2 states associated with GMM 3 Gaussians, combined LWE (Log decomposition based Energy) descriptor Coiflet wavelet mother level 4. A comparative state art methods shows efficiency our approach that reduces mean factor close healthy subjects 1.3 subjects.