0
Research Papers

An EMG-Driven Biomechanical Model That Accounts for the Decrease in Moment Generation Capacity During a Dynamic Fatigued Condition

[+] Author and Article Information
Guillaume Rao1

Institute of Movement Sciences, University of the Mediterranean, UMR CNRS 6233, 163, Avenue de Luminy, 13288 Marseille Cedex 09, France; Department of Mechanical Engineering, University of Delaware, 126 Spencer Laboratories, Newark, DE 19716guillaume.rao@univmed.fr

Eric Berton

Institute of Movement Sciences, University of the Mediterranean, UMR CNRS 6233, 163, Avenue de Luminy, 13288 Marseille Cedex 09, Franceeric.berton@univmed.fr

David Amarantini

Universite de Toulouse, UPS, LAPMA, 118, Route de Narbonne, F-31062 Toulouse Cedex 09, Francedavid.amarantini@cict.fr

Laurent Vigouroux

Institute of Movement Sciences, University of the Mediterranean, UMR CNRS 6233, 163, Avenue de Luminy, 13288 Marseille Cedex 09, Francelaurent.vigouroux@univmed.fr

Thomas S. Buchanan

Department of Mechanical Engineering, University of Delaware, 126 Spencer Laboratories, Newark, DE 19716buchanan@udel.edu

1

Corresponding author.

J Biomech Eng 132(7), 071003 (May 14, 2010) (9 pages) doi:10.1115/1.4001383 History: Received February 11, 2008; Revised April 14, 2009; Posted March 08, 2010; Published May 14, 2010; Online May 14, 2010

Abstract

Although it is well known that fatigue can greatly reduce muscle forces, it is not generally included in biomechanical models. The aim of the present study was to develop an electromyographic-driven (EMG-driven) biomechanical model to estimate the contributions of flexor and extensor muscle groups to the net joint moment during a nonisokinetic functional movement (squat exercise) performed in nonfatigued and in fatigued conditions. A methodology that aims at balancing the decreased muscle moment production capacity following fatigue was developed. During an isometric fatigue session, a linear regression was created linking the decrease in force production capacity of the muscle (normalized force/EMG ratio) to the EMG mean frequency. Using the decrease in mean frequency estimated through wavelet transforms between dynamic squats performed before and after the fatigue session as input to the previous linear regression, a coefficient accounting for the presence of fatigue in the quadriceps group was computed. This coefficient was used to constrain the moment production capacity of the fatigued muscle group within an EMG-driven optimization model dedicated to estimate the contributions of the knee flexor and extensor muscle groups to the net joint moment. During squats, our results showed significant increases in the EMG amplitudes with fatigue ($+23.27%$ in average) while the outputs of the EMG-driven model were similar. The modifications of the EMG amplitudes following fatigue were successfully taken into account while estimating the contributions of the flexor and extensor muscle groups to the net joint moment. These results demonstrated that the new procedure was able to estimate the decrease in moment production capacity of the fatigued muscle group.

<>

Figures

Figure 1

Typical posture of a subject during the MVC and the isometric fatigue tasks

Figure 2

Schematic representation of the methodology used in the study to assess the contributions of the flexor and extensor muscle groups to the net joint moment under fatigue condition

Figure 3

Schematic representation of the methodology used to estimate the frequency content of the EMG signal. Upper graph shows the evolution through time of the knee angle. Middle graph represents the EMG signal of the rectus femoris muscle. Bottom graphs show the outputs of the wavelet analysis. The 512 points analysis windows were centered at the time instants were the knee joint angle trajectory was the most repeatable. For the bottom graphs, the time is along the horizontal axis and the frequency along the vertical one. The amplitude of the signal at any given time and frequency is represented using color scale. Brighter colors inform that the corresponding frequency bandwidth is often found in the signal frequency content.

Figure 4

Representative example for one subject of the isometric linear relationship that links the force production capacity of the rectus femoris muscle (normalized force/EMG ratio) to the normalized mean frequency decrease. Raw data as well as the corresponding linear regression are shown.

Figure 5

Evolution through time of the mean values (n=9) of the moments developed by the extensor muscle group (upper curves), the net joint moments (middle curves) and the flexor muscle group moments (lower curves). For each variable, thick lines correspond to the nonfatigued condition and thin lines to the fatigued situation.

Figure 6

Evolution through time of the individual muscle gains of the rectus femoris (thick lines) and vastus medialis (thin lines) muscles in the nonfatigued (continuous lines) and fatigued (dashed lines) situations. Note the similar patterns of the wi(t) coefficients, which imply that the optimization procedure did not take advantage of this variable to compensate for the additional equality constraint.

Errata

Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Related Proceedings Articles
Related eBook Content
Topic Collections