Probability density of electromyography signal for different levels of contraction of biceps brachii

Abstract

The probability density function (PDF) of surface electromyography (sEMG) signals can be modelled with the Gaussian and Laplacian PDFs. However, the sEMG PDF is dependent on the levels of contraction of the muscles. Different techniques have been proposed for testing Gaussianity levels of sEMG, i.e., kurtosis, negentropy, and mean bicoherence power, whereas the suitable technique has not been reported yet. In this paper, the experimental sEMG PDF and its relationship with the levels of muscle contraction were re-examined and the suitable Gaussianity test was presented based on sEMG acquired from biceps brachii muscle during the static contraction of the muscle at different load levels. The results show that the EMG PDF was non-Gaussian at low level of contraction and it tends to be more Gaussian at high level of contraction. Kurtosis and negentropy decreased as muscle load levels increased, whereas the consistent results between mean bicoherence power and muscle load levels were not found. Negentropy showed a better linear relationship with the levels of contraction of biceps brachii than other techniques.