A Meshfree Boundary Method for M/EEG Forward Computations

Abstract

The brain activity can be investigated in a noninva sive way by means of electromagnetic techniques, i.e., electroen cephalography (EEG) and magnetoencephalography (MEG), which requires a typical inverse problem to be solved. Therefore, an accurate and fast forward solver is needed. As a truly meshfree alternative to the Boundary Element Method (BEM), we propose the Method of Fundamental Solutions (MFS) for solving the M/EEG forward problem. The solution of the forward problem is obtained, via the ethod of Particular Solutions (MPS), by numerically solving a set of coupled boundary value problems for the Laplace equation. By investigating both numerical accuracy and computational load for spherical geometries and comparing them with a state-of-the-art BEM solver, we show that the proposed method is attractive