EEG inverse problem solution with minimal influence of the conductivity

In this paper, we propose a novel method that improves the accuracy of the estimation of neural electrical dipoles when solving the EEG inverse problem. A spherical head model is used where we limit the influence of the unknown conductivity brain-skull ratio on the inverse problem. We redefine the cost function that is used in the EEG problem where only useful information is used as input in the inverse problem. In contrast to previous approaches, weighting factors are used where the electrodes are strategically chosen so to reduce the error made on EEG dipole source localization. The proposed method enhances the source localization accuracy from approximately 9mm to 1mm for dipoles near the edge and from 2.1mm to 0.4mm for dipoles near the center of the brain.