Relative influence of model assumptions and measurement procedures in the analysis of the MEG.

Abstract

The relative influences of several model parameters and measurement setups on the MEG are studied quantitatively. The influences of the number of grid points of the flux transformer and the accuracy of the measurements are analysed using as criterion a relative difference measure (RDM). In a similar way, using the RDM, the influence of various models is evaluated. The volume conductor, i.e. the head, is described by three different compartment models: the first model consists of concentric spheres, the second of eccentric spheres optimally fitting the individual compartments of the head, and the third consists of realistically shaped compartments. The evaluation of the influence of the model of the source on the MEG is studied by taking either one single current dipole or a set of two current dipoles. The RDM described in the paper is shown to be a valuable measure in the quantitative analysis of MEGs.