Muscle fatigue detection through wearable sensors: a comparative study using the myo armband

Abstract

Novel wearable systems allow the measure of very complex physiological phenomena extending their capabilities and maintaining their non-invasiveness. A good example of this is the use of superficial electrodes for recording electromyography signals (also called superficial electromyography- sEMG) which can reveal information regarding muscle force and fatigue. Aiming at demonstrate the accuracy of a commercial grade wearable system for sEMG, the Myo Armband for fatigue measurement, we carried out a comparative study. 3 subjects were used under a standard protocol for fatigue detection using two different sensors: a Base ground-truth sEMG sensor, and the commercial wristband Myo, both connected in the biceps brachii. Time and frequency domain parameters were compared using an ANOVA test and a correlation analysis. Results showed a median correlation for the three subjects between 0.4 and 0.6 between the Base Sensor and the Myo Armband signals exposing significant differences p <0.05 for all three cases. The biomarkers of the sEMG signal of both sensors were consistent research found in the literature. Novel wearables sensors can be used in medical scenarios where high accuracy is not a requirement, instead, non-invasiveness can provide ubiquity for rehabilitation treatments as well as a continuous signal recording and data logging processes.