Novel Feature Modelling the Prediction and Detection of sEMG Muscle Fatigue towards an Automated Wearable System

Surface Electromyography (sEMG) activity of the biceps muscle was recorded from ten subjects performing isometric contraction until fatigue. A novel feature (1D spectro_std) was used to extract the feature that modeled three classes of fatigue, which enabled the prediction and detection of fatigue. Initial results of class separation were encouraging, discriminating between the three classes of fatigue, a longitudinal classification on Non-Fatigue and Transition-to-Fatigue shows 81.58% correct classification with accuracy 0.74 of correct predictions while the longitudinal classification on Transition-to-Fatigue and Fatigue showed lower average correct classification of 66.51% with a positive classification accuracy 0.73 of correct prediction. Comparison of the 1D spectro_std with other sEMG fatigue features on the same dataset show a significant improvement in classification, where results show a significant 20.58% (p < 0.01) improvement when using the 1D spectro_std to classify Non-Fatigue and Transition-to-Fatigue. In classifying Transition-to-Fatigue and Fatigue results also show a significant improvement over the other features giving 8.14% (p < 0.05) on average of all compared features.

Abstract licence: CC BY

• A Review of Non-Invasive Techniques to Detect and Predict Localised Muscle Fatigue · 2011

A person counts once a day per paper. We keep no cookie and no address: a reader is a hash that changes every day. Any month under five reads shows no number.