Identifying the Motor Neuron Disease in EMG Signal Using Time and Frequency Domain Features with Comparison

Motor neuron diseases are the most common neurological disorders found in the age ranges between 35-70years, which selectively affect the motor neurons. Amyotrophic lateral sclerosis (ALS) is a fatal motorneuron disease that assails the nerve cells in the brain. This disease progressively degenerates the motorcells in the brain and spinal cord, which are responsible for controlling the muscles that enable human tomove around, breathe, speak, and swallow. The electromyography (EMG) signals are the biomedicalsignals that are used to study the muscle function based on the electrical signal originated from themuscles. As the nervous system controls the muscle activity, the EMG signals can be viewed and analyzedin order to detect the indispensable features of the ALS disease in individuals. In this paper, analyzing thetime and frequency domain behaviour of the EMG signals obtained from several normal persons and theALS patients, some characteristic features, such as autocorrelation, zero crossing rate and Fouriertransform are proposed to identify the ALS disease. For the pupose of classification, K-nearestneighbothood classifier is employed in a leave-one out cross validation technique. In order to show theclassification performance, an EMG database consisted of 7 normal subjects aged 21-37 years and 6 ALSpatients aged 35-67 years is considered and it is found that the proposed method is capable of distinctlyseparating the ALS patients from the normal persons.