Hand grip force estimation via EMG imaging

Abstract

Physiological muscle disorders resulting from various diseases, such as sarcopenia or strokes, affect millions of people each year according to the American Heart Association. Medical examinations for assessing muscle health are difficult to administer and lack standardization. Surface electromyography (sEMG) signals have found a wide variety of applications; in this work, we present a simple reliable framework for hand grip force estimation using them. We hope that this work will lay the groundwork for applying hand grip force estimation as a method for standardized and reliable muscle health assessment. Seven healthy male subjects were voluntarily recruited and sEMG signals were collected from eight electrodes uniformly distributed around the forearm. A logarithmic function can describe the EMG–force relationship. We propose a novel method for generating functional potential-based inverse EMG images of the forearm. The hand grip force can be estimated from the reconstructed images. Using a simple lightweight system and simple algorithmic techniques we can attain a mean correlation coefficient of 0.95±0.01 and mean RMSE of 0.18±0.08N. The results are presented for subject-dependent tests, the generalizability of the method is still to be researched over a larger cohort of subjects, including rehabilitation patients.