Developing a Fuzzy C-Means Inference System for Muscle Strength Prediction Using sEMG

The research involved the development of a fuzzy inference system (FIS) to predict grip strength through the analysis of sEMG signals. Fuzzy rules were generated using fuzzy c-means (FCM) clustering. In an experiment focused on lifting motions and including nine participants, the FIS demonstrated promising results. Specifically, in the non-weight condition, the FIS achieved a Mean Squared Error (MSE) of 0.1246 and an R-squared value of 0.3357 for grip strength prediction. However, when an 11-lb weight load was introduced, the accuracy of the FIS dropped significantly, leading to less reliable predictions. This was reflected in an increased MSE of 0.1804 and a reduced R-squared value of 0.0379. These outcomes underscore the potential of wearable sEMG devices coupled with a fuzzy inference system for grip strength prediction. The research also highlights the need for further research in this evolving field.