Single dipole source localization from conventional EEG using BP neural networks

Abstract

Brain source localization is an important inverse problem for brain diagnosis and functional analysis. The usefulness of BPNN (back propagation neural network) to solve this problem has been reported. The purpose of the study is to examine the practical effectiveness of NN for single dipole source localization from EEG based on conventional 18 electrodes of a 10-20 system, and to investigate estimation accuracy in relation to EEG references, dipole positions and network training.