Modelling Artificial Stimulation and Response in Peripheral Nerves Including Ephaptic Interactions

This research aims to (1) extend our knowledge on the response of peripheral nerves to artificial stimulation for sensory feedback provision from neural interfaces, and (2) create a computational tool to facilitate this study. We were interested in studying how ephaptic coupling between myelinated fibers influences activity in nerve trunks under artificial stimulation and during action potential propagation. Ephaptic interaction simulations in nerve trunks were performed to quantify this influence. For this, we created peripheral nerve models containing electrodes for electrical stimulation and recording within a tool that can be further used in electrode design optimisation and neural activity research. The created model can use a self-contained or a hybrid field-neuron method. The self-contained method uses a resistor network that electrically couples all axons, tissues, electrodes, and surrounding medium, and is solved by the NEURON simulation environment. The resistor network uses weighted Voronoi tessellations in the Laguerre geometry to define the electrical connections between all nerve elements given any cross-sectional anatomy. The hybrid field-neuron approach also uses the resistor network to compute the fields, but uses them stimulate fiber in a separate simulation. The self-contained model was designed so that it could simulate artificial stimulation, neural activity with ephaptic coupling and electrode recordings simultaneously. Researchers often assume ephaptic coupling is weak among myelinated axons, and therefore, tend to ignore it. Simulations carried out in this work, however, show that ephaptic coupling increases axon recruitment during artificial stimulation. This effect should be taken into account in further research. On the other hand, ephaptic coupling during propagation in realistic bundles with large numbers of heterogeneous myelinated fibers is weaker, unstable, and more complex than what is known from previous studies on bundles of few homogeneous fibers. This research provides detailed results and insights on these aspects of peripheral neural activity.