Three-dimensional model of a muscle and simulation of its surface EMG

The aim of this paper was to present a spatial model of a muscle including all its motor units (MU) and a simulation of its surface EMG. The simulations are part of a larger model including in addition the input system to the motoneuronal pool, the motoneuronal pool itself and the force generating mechanism. The muscle and the MU territories are represented by elliptic cylinders. Two algorithms ate presented to position the MU territories within the muscle. The final goal was to achieve a final global fiber density, which is as constant as possible. The algorithm, which minimizes the variability of the fiber density each time a MU territory is positioned, proved to be superior. The surface EMG of this model muscle was simulated by assuming that each muscle fiber generates an action potential (AP) at the motor endplate in the middle of the fiber and propagates it at constant velocity to both ends. APs were represented by a tripole and the sum of the potentials evoked by the tripoles generates a fiber AP at the recording site. All the fibers within the MU territory generate the MU AP and finally all active MUs together give rise to the surface EMG. As an example, the steady activity of the human first dorsal interosseus muscle was simulated. The surface EMGs, recorded with an array of electrodes around and along the muscle, were illustrated.