Modeling and Simulation of EMG Signal and H-Reflex in Healthy Human Subject

Abstract

Electromyography (EMG) is a signal representing the activity of skeletal muscles. Besides, H-reflex and M-wave are two responses to the electrical stimulation of the nerve, and they appear in EMG. In this paper, the purpose was to model the recorded EMG from the FCR muscle in response to both the electrical stimulation and the brain commands during motor imagery. This model could capture the qualitative behavior of the H-reflex and how physiological variables change its waveform. Two mathematical models of the EMG signal (i.e., baseline EMG and H-reflex) were developed. These models were designed to simulate the EMG signal with respect to the intensity of contraction and electrical stimulation. Also, the effect of the brain descending pathways was considered with a parameter. We used the Hodgkin-Huxley (HH) and Morris-Lecar (ML) equations for the action potentials (APs) of motor neurons and muscle fibers, respectively. After processing the sensory signals in the spinal cord, the response pulses are transmitted from the spinal cord to the muscle. When the AP is generated, the spike train passes to the muscle fiber. Each muscle contains several motor units (MUs), each of which consists of a motor neuron and its innervated muscle fibers. Moreover, summation of the total activity of the muscle fibers within each MU forms its activity. Likewise, total muscle activity (i.e., EMG) is the summation of the activity of all MUs. The H-reflex and M-wave models were developed by applying electrical stimulation in the EMG model. Moreover, mean square error (MSE) for the model was 0.00027%, and Spearman`s correlation coefficient was 0.722 (p-value < 0.05).