Volume conduction models for surface EMG; confrontation with measurements

Abstract

Volume conduction models are used to describe and explain recorded motor unit potentials (MUPs). So far it has remained unclear which factors have to be taken into account in a volume conduction model. In the present study, five different models are confronted with measured MUP distributions over the skin surface above the m. biceps brachii generated by MUs at different depths and recorded by small surface electrodes. All model simulations include fibres of finite length. The models differ in the size of the volume conductor (finite/infinite), the number of different layers (1, 2 or 3) and the conductivities of these layers (representing muscle, subcutaneous fat and skin). All measured and simulated MUPs contain a mainly negative propagating wave followed by a positive wave simultaneously present at all electrode positions. The magnitude of the different MUP components relative to each other and as a function of motor unit (MU) and electrode position differ between the models studied and the measurements. All simulated MUPs changed faster with observation distance than the measured MUPs. The three-layer model, in which muscle tissue was surrounded by a subcutaneous fat layer and by a layer of skin resulted in MUPs closest to the measured MUPs.