Evaluation of different EEG source localization methods using testing localization errors

Abstract

The ideas underlying the quantitative localization of the sources of the EEG review within the brain along with the current and emerging approaches to the problem. The ideas mentioned consist of distributed and dipolar source models and head models ranging from the spherical to the more realistic based on the boundary and finite elements. The forward and inverse problems in electroencephalography will debate. The inverse problem has non-uniqueness property in nature. More precisely, different combinations of sources can produce similar potential fields occur on the head. In contrast, the forward problem does have a unique solution. The forward problem calculates the potential field at the scalp from known source locations, source strengths and conductivity in the head, and it can be used to solve the inverse problem. In the final part of this paper, we compare the performance of three well–known EEG source localization techniques which applied to the underdetermined (distributed) source localization of the inverse problem. These techniques consist of LORETA, WMN and MN, which comparing by testing localization error