Classification of normal, ALS, and myopathy EMG signals using ELM classifier

Electromyography (EMG) signals collected from skeletal muscles, can be used for diagnosis of ALS and myopathy diseases. This paper presents a new technique based on five features and Extreme learning machine (ELM) classifier for the diagnosis of ALS and myopathy diseases. Bispectrum based features with mean plosion index, and fractal dimension features are extracted from EMG signals. For the purpose of automatic multi-class EMG signal classification, a developed learning algorithm ELM has been used. Experimental results of EMG data set show the improved performance of proposed method, and it is evident that the proposed method can be effectively used for the diagnosis of ALS and myopathy diseases.