Methods for Localization of Electrical Sources in the Human Brain and Applications to the Visual System

Author: Darcey, Terrance Michael

Year: 1979

Degree: Dissertation (Ph.D.)

Advisor: Fender, Derek H.

Committee Member: Unknown, Unknown

Option: Engineering

DOI: 10.7907/yvch-jm69

Abstract

Multichannel human visual evoked scalp potential (VESP) recording was combined with results from electric field theory and the methods of non-linear parameter estimation to see if it was possible to infer the general spatio-temporal course of brain excitation in the processing of a visual stimulus, To do this, various models of the head and underlying sources were tested , Head models included a simple homogeneous and 2 different piecewise homogeneous volume-conducting spheres, the latter of which were designed to account for the known electrical inhomogeneity and anisotropy provided by the skull and scalp. Source models studied were single and multiple electrostatic dipoles and two simple types of extended sources, radial cap and annular sector dipole layers. Equivalent sources for the experimental data were found by taking least-squares estimates of model parameters, minimizing the sum of the squared deviations between the actual scalp potentials and the scalp potentials computed by the model, Since these models were parameterized for both location and source characteristics (e.g. orientation and magnitude), one could attempt to relate them to the topography and functional anatomy of the brain. A working assumption in relating these was that active cortex could be modeled to a first approximation as a neuronal layer wherein the neurons are treated as discrete current sources oriented normal or at right-angles to the cortical surface, Thus, it was assumed that VESP's reflect the topography of the tissues that produce them.

40-channel VESP measurements were made to briefly appearing checkerboard patterns placed in various areas of the visual field. These data were analyzed in terms of underlying equivalent sources in the brain. Although there were some variations present in the 3 subjects studied, these may be consistent with known variations in cortical topography. The measured potential distributions showed a radical dependence of VESP topography on stimulus locus and indicated that these VESP's are probably volume-conducted field effects arising from a small number of fairly localized sources in the brain. Equivalent dipoles in general give excellent fits to the measured data and the mapping between the visual field and these equivalent sources is similar to the connnonly accepted mapping between the visual field and visual cortex. As such, it appears that evoked potentials may be used as a crude tool for defining the anatomical projection of localized sensory fields onto the cortical surface in humans.

Files