Analysis of surface electromyographic parameters for the assessment of muscle fatigue during moderate exercises

Abstract

[EN] Muscle fatigue is a neuromuscular phenomenon which prevents the muscles to generate the required force. Although there are multiple studies about detecting this condition with surface electromyography, this method has still some limitations. The objective of this study was to determine the viability of some biomarkers as early indicators of fatigue during moderate exercise. Eight healthy volunteers performed 2 dynamic exercises involving the flexo-extension of the knee lifting 2 and 4kg attached to the ankle. A bipolar electromyographic signal was recorded simultaneously from: Rectus Femoris, Vastus Lateralis and Biceps Femoris. Muscle activations were extracted by supervised automatic segmentation and their root mean square, bandwidth, entropy and spectral moment were calculated. The maximum of the normalized coefficient of the linear cross-correlation between each muscle pair were also computed as well as the slope throughout the execution of the exercise repetitions of all above mentioned parameters. Results from 2kg exercise did not reveal clear signs of fatigue. By contrast those from 4kg showed an increase of the root mean square and spectral moment ratio, and decrease of bandwidth, sample entropy and linear cross correlation coefficient, pointing out fatigue. In this study, it has been confirmed the viability of the cross-correlation technique as a possible new biomarker related to the first signs of muscle fatigue.