Topography Estimation of Visual Evoked Potentials Using a Combination of Mathematical Models

Abstract

This study proposes a method for estimating the topographical distribution of Visual Evoked Potentials (VEPs) from separated power spectrum components by a combination of models. VEPs with various temporal frequencies were recorded from nine healthy adults. The original power spectrum consisted of the VEP; background activities, artifacts, and other components were then obtained. To extract the VEP components from the original power spectrum, models corresponding to background activities, especially for posterior alpha rhythm, the low-frequency component and the high-frequency component, caused due to the EMG artifact, were constructed, and the relevant parameters were estimated. Finally, VEP components were calculated by subtracting them from the original power spectrum. The topographical distribution of the first harmonic (1F) and second harmonic (2F) components of the VEP were obtained by the proposed method. The estimation of the other components, aside from the VEPs, was also investigated. The merits and usefulness of the proposed method were analyzed with a comparison to the conventional stimulus-locked averaging method. The proposed method has several advantageous points compared to the conventional averaging method. Specifically, the posterior alpha rhythm and the EMG artifact were accounted for directly in the estimation of the VEP components. Therefore, an accurate estimation of the VEP components can be performed even the measurement of the components are prone to the error.