A simulation study on spatial filters for cortical source imaging from EEG

The objective of this study was to explore suitable spatial filters for inverse estimation of cortical equivalent dipole layer from the scalp EEG. The performance of the Weiner filter and the projection filter as a restoration filter was examined by computer simulations. The parametric Weiner filter (PWF) and the parametric projection filter (PPF) were applied to an idealized 3D head model under the various noise conditions in order to estimate the dipole layer distribution from scalp potentials. The PPF and PWF algorithms were compared with commonly used inverse regularization techniques, such as the general inverse using the truncated singular value decomposition (TSVD) and the Tikhonov regularization (TKNV). The present simulation results suggest that the PWF and PPF perform better than that of the TSVD and the TKNV in the condition of edge-concentrated non-uniform noise when the correlation between the signal and noise is low.