Peripheral Nerve Stimulation Modeling for MRI

Rapid switching of magnetic fields in the human body induces electric fields strong enough to induce peripheral nerve stimulation (PNS). The occurrence of PNS often limits the usable performance of the MR gradient system, leading to longer scan times or reduced spatial and temporal resolution. Despite the impact of PNS on the scanning performance, the tools to fully predict stimulation locations and thresholds have only recently been developed. These advances utilize improved modeling of the induced electromagnetic fields within detailed body models, including the field magnitude and orientation with respect to major peripheral nerves. Combining this with detailed neurodynamic models then enables the prediction of stimulation thresholds and their locations within the body. It also provides a window for evaluating the contributions of the components of the gradient winding pattern as well as capturing the internal dynamics of the nerve during the applied field modulation. Together, this will hopefully prove valuable in the development of PNS mitigation strategies. This article reviews recent (and less recent) approaches in electromagnetic and neurodynamic modeling and describes how these modeling disciplines can be coupled to understand and predict gradient-induced action potential generation within the peripheral nervous system. The potential use of these PNS prediction models is illustrated by characterizing and comparing the PNS capabilities of two head-insert MR gradient coils.