Modeling the Recording of Intraneural Action Potentials with Microelectrodes Using FEM and Point-Source Methods.

To better serve the demands of the amputee population, there is a need to increase the number of independent channels of motor-control and sensory-feedback information used with prosthetic limbs. As a result, there is a drive to increase the number of channels in intraneural peripheral-nerve interfaces. Computational modeling can inform efforts to improve the design of neural interfaces. We compare two coupled modeling methodologies for recording action potentials using a biophysical model tuned for recording in mammalian motor nerve fibers: the analytical point-source model and a finite-element-method (FEM) reciprocity model. After assessing the size of the model space needed for high accuracy, the two methods are compared. Results show the models agree well and reveal the effect of electrode size, fiber-electrode separation, and fiber diameter on recorded action potentials.