Characterizing the Dynamics of Surface Electromyography Signals in Muscle Fatigue Through Visibility Motif Networks

Monitoring of the physiological function during exercise can provide insights on the quality of the training and prevent injury. Specifically, the signals from the muscle sensors (surface electromyography) are difficult to interpret and limited attempts have been made to develop effective algorithms for the real-time monitoring of muscle fatigue. In this work, the applicability of visibility graph motif features for the real-time monitoring of muscle fatigue is explored. Experimental investigations have been conducted on 58 healthy adult volunteers. Results indicate that the network entropy features are able to characterize the changes in signal dynamics in nonfatigue and fatigue conditions. These metrics have the potential to be used as a marker to predict functional capabilities of humans in real-world scenarios.