Impact of Axon Model Complexity on Deep Brain Stimulation: A Comparative Analysis of MRG and Cohen Double-Cable Models

Deep brain stimulation (DBS) relies on biophysical neuron models to predict neural activation and stimulation spread. This study compared two axon models, the McIntyre-Richardson-Grill (MRG) and Cohen models, to evaluate how structural and electrophysiological differences impact DBS outcomes. Both models were simulated using extracellular electric field stimuli from 2202 DBS leads and subjected to biphasic pulse stimulation with varying parameters.

Activation distances ranged from 2 to 10 mm, with Model II showing greater excitability at a given current. Threshold differences ranged from -1.40 mA to 0.27 mA, with Model II consistently requiring lower thresholds. Both models exhibited an inverse relationship between pulse width and activation threshold, but Model II showed increasing thresholds at higher frequencies, while Model I slightly decreased.

Machine learning models trained on distance, pulse width, and frequency achieved high predictive accuracy, with Model II performing best. Both models reliably estimate DBS-induced activation, but structural differences affect excitability and frequency-dependent behavior. With awareness of their strengths and limitations, either model can estimate activation distances for predicting electric field isolevels in patient-specific DBS simulations.