Accurate localization of electrical sources in the brain using EEG's or MEG's would be a very valuable clinical and research tool. At present, accuracy is poor, in part, because the effects of the skull and head shape on EEG's and MEG's and source localization are not adequately known. In addition, little or no information is available about the effects that the distance from the head at which MEG's are measured has on these measurements and on localization. Particularly needed is information about whether MEG's measured at some fixed distance from the surface of the head or on the smallest spherical surface surrounding the head provide the most accurate localization. Information about the effects of distance on MEG's is also needed for the new generation of multichannel MEG detectors that is coming into use. This research would use computer modeling methods to determine the effects of these various factors.
The specific aims of this research are: (1) to develop a computer model of the human head and to experimentally verify that it is a good representation of the head; (2) to use this model to determine the effects of the skull and head shape on EEG's and MEG's and to compare the effects on the two types of measurements; (3) to use this model to determine the effects of the skull and head shape on source localization and compare the effects on localizations using the two types of measurements; and (4) to use this model to determine the effects that the distance from the head at which MEG's are measured has on these measurements and on localizations using them. The head model would be capable of being easily modified to represent various features of the skull and head shape such as local variations in skull thickness, conductivity, and/or shape, the thick, bony structures that form the base of the skull, the irregular shape of the head, etc. Hence, the effects of a wide range of and various combinations of features of the skull and head shape on EEG's and MEG's and source localization could be easily determined using this model. Information concerning these effects is needed to improve source localization accuracy. The model would also be used to obtain information on the effects of measurement distance on MEG's. This information is required by MEG researchers to design the most accurate MEG measurement systems and detectors.
