A Predictive model for the detection of Muscle fatigue using sEMG signal

Abstract

Surface electromyography (sEMG) is an important dimension for analyzing exercise and muscle activity. sEMG requires a very high sampling rate, thus wireless transmission of generated signals becomes very challenging. An important application of sEMG monitoring is the detection of muscle fatigue. The present study proposes a novel framework for the detection of muscular fatigue by monitoring sEMG signals obtained from various muscle groups throughout the body. The system uses an LSTM predictive model for the binary classification of sEMG signals trained on the UCI dataset.