Cortical Dipole Imaging of Movement-related Potentials by Means of Parametric Inverse Filters Incorporating with Signal and Noise Covariance

Abstract

The objective of this study is to explore suitable spatial filters for inverse estimation of cortical equivalent dipole layer imaging from the scalp electroencephalogram. We utilize cortical dipole source imaging to locate the possible generators of scalp-measured movement-related potentials (MRPs) in human. The effects of incorporating signal and noise covariance into inverse procedures were examined by computer simulations and experimental study. The parametric projection filter (PPF) and parametric Weiner filter (PWF) were applied to an inhomogeneous three-sphere head model under various noise conditions. The present simulation results suggest that the PWF incorporating signal information provides better cortical dipole layer imaging results than the PPF and Tikhonov regularization under the condition of moderate and high correlation between signal and noise distributions. On the other hand, the PPF has better performance than other inverse filters under the condition of low correlation between signal and noise distributions. The proposed methods were applied to self-paced MRPs in order to identify the anatomic substrate locations of neural generators. The dipole layer distributions estimated by means of PPF are well-localized as compared with blurred scalp potential maps and dipole layer distribution estimated by Tikhonov regularization. The proposed methods demonstrated that the contralateral premotor cortex was preponderantly activated in relation to movement performance. In cortical dipole source imaging, the PWF has better performance especially when the correlation between the signal and noise is high. The proposed inverse method was applicable to human experiments of MRPs if the signal and noise covariances were obtained.