Detection of muscle fatigue using wearable (MYO) surface electromyography based control device

Wearable technology had set to play a major role in solving various problems in medical or sports applications in sports training, ergonomics, and mostly physical therapy. Surface electromyography (sEMG) has been used for localised muscle fatigue research area, although the success of sEMG based techniques is currently limited to isometric contraction and is not acceptable to the human movement community. This work propose the use of sEMG based sensor to measure the muscle activity during dynamic contractions. In this study, the use of a low cost commercial available sensor called MYO armband was implemented to reveal the viability to utilise this type of sensor to quantify muscle fatigue. The acquired sEMG signals from the muscle were analysed with the use of a set of selected features, RMS and MDF have been used as indicators of fatigue, and then classifying the signal into two classes non-fatigue and fatigue using backpropagation neural networks, allowing the prediction of muscle fatigue. Results show that automating the process of localised muscle fatigue detection is showed with higher accuracy compared with other methods.