IMPROVED FORWARD EEG CALCULATIONS USING LOCAL MESH REFINEMENT OF REALISTIC HEAD GEOMETRIES

A method for semi-automatically constructing realistic surface meshes of 3 head structures - scalp, skull and brain - from a stack of MR ima ges is described. Then an evaluation is given for both spherical and r ealistic dipolar models, using the boundary element method (BEM). In b oth cases, locally refined models were considered. Two characteristic mesh parameters were defined: the global and the local mesh densities (in triangles per cm(2)). In spherical geometries, numerical and analy tical solutions were compared, and in the realistic case, all models w ere compared to a highly refined one, considered as a reference. Both geometries gave comparable results. It was found that for ''deep dipol es'' located at more than 20-30 mm under the brain surface, meshes wit h a global density of 0.5 tri/cm(2) gave ''acceptable'' results, where as for more superficial dipoles (2-3 mm < depth < 20-30 mm), it was ne cessary to locally refine meshes near the source location up to a loca l density of about 5-8 tri/cm(2), to get comparable results.