LOCALIZING BRAIN ACTIVITY WITH VEP'S

点击翻译按钮获取中文摘要

DESCRIPTION (Investigator's abstract): This project addresses the problem of localizing brain activity in humans from visually evoked potential fields measured on the scalp. Previous work has set the localization problems in terms of defining brain current sources usually in the form of current dipoles. This leads to a problem that is ill posed, i.e., no unique solution is possible and to an undesirable linkage between the localization problem and specific models of how current is generated in the brain. One approach to ill-posed problems is to solve a closely related problem that provides the needed information but which has a unique solution. Brain surface potentials are uniquely specified by scalp potential fields and localize brain activity with good resolution. No assumptions about brain electrogenesis are required. Computation of brain surface potentials fields from scalp potential fields does require quantitative anatomy of the scalp and brain and accurate measurements of electrode position. Techniques to accomplish this using a 3-D digitizer and magnetic resonance imaging of the brian are described. Verification is proposed by means of a life-like model of the human head based on a human skull and by mapping the retino-topic projections to striate cortex in humans. The ability to localize brian activity with VEP's would provide a very useful tool for basic and clinical research in vision because the temporal resolution of VEP's is much better than that of position emission tomography, the principal method now available for regional brain activity measurement. As such it could provide the means to track brain activity as it proceeds in time after a stimulus is presented. This could be useful in understanding normal visual processing, in defining abnormal processing, and in studying adaptations to brain injury.