Preprocessing General Head Models for BEM-FMM Modeling Pertinent to Brain Stimulation

Abstract

Introduction: Transcranial magnetic stimulation (TMS) is a major noninvasive neurostimulation method in which a coil placed near the head employs electromagnetic induction to produce electric fields and currents within the brain. To predict the actual site of stimulation, numerical simulation of the electric fields within the head using high-resolution subject-specific head models is required. A TMS modeling software toolkit has been developed based on the boundary element fast multipole method (BEM-FMM), which has several advantages over conventional finite element method (FEM) solvers. Objective: To extend the applicability of the BEM-FMM TMS simulation toolkit to head models whose meshing scheme produces a single mesh for every unique tissue instead of producing a single mesh for every unique tissue/tissue boundary. Method: The MIDA model of the IT’IS Foundation, Switzerland, comprises 115 high-resolution tissue models in the form that the BEM-FMM toolkit is modified to accept. The updated BEM-FMM toolkit is tested using this head model. Results: The BEM-FMM toolkit has been successfully modified to accept head models consisting of one unique mesh per unique tissue while still supporting its initial model format of one unique mesh per boundary between two specific tissues. Performance impacts occur in the preprocessing phase only, meaning that the charge computation method performs equally well regardless of model format.