Modeling Surface Electrode Stimulation

Student Author(s)

Kathleen Finn
Jessica Gaines

Faculty Mentor(s)

Dr. Katharine Polasek

Document Type


Event Date



Phantom limb pain is experienced by 50-80% of amputees. We hypothesize that eliciting a “real” sensation in the phantom limb using surface electrode stimulation may reduce or eliminate phantom limb pain. To assist in eventual electrode placement, a computer model is being developed to predict the effect of electrode location, size, and configuration on median nerve activation. A three-dimensional finite element model of the elbow was created with ANSYS Maxwell, using two illustrated cross sections of the arm swept together. Simulations were performed by applying a voltage across two 17x30 mm electrodes placed over the elbow. The median nerve was modeled with 10 fascicles in three random arrangements and NEURON was used to predict axon firing. A sensitivity analysis was performed to determine the effect of the number of nodes per axon, number of axons per fascicle, and resolution of exported voltages on nerve firing. After investigation of these parameters, the model can predict how many motor axons fire for a given fascicle size, fascicle location, and stimulation value. The next step will be performing a similar analysis with sensory axons. To adapt the model to account for the difference between the channel properties of motor and sensory axons, a sensory axon model was developed based on a motor axon model by McIntyre, Richardson, and Grill (2002) and information from literature. Parameters were adjusted to reflect the differences in channel properties. A sensitivity analysis was conducted to find the parameters with the largest effect on threshold, and these significant parameters were investigated further. The models are still in progress but presently predict a slightly lower threshold voltage for a sensory axon than a motor axon at a given fiber diameter. This differs from the predictions of the original model but is closer to experimental observations.


This research was supported by the Hope College Dean Start-up Fund, the Christine Tempas Engineering Summer Research Fund¸ the Michigan Space Grant Consortium, and an award to Hope College from the Howard Hughes Medical Institute through the Undergraduate Science Education Program.

This document is currently not available here.