A mixed discretized adjoint double layer formulation for the electroencephalography forward problem with high brain-skull contrast ratios

The Electroencephalography (EEG) forward problem is of key importance for EEG-based neuroimaging. However, it is traditionally plagued by computational inefficiencies every time the subject model's skin-to-skull conductivity ratio is high. In this paper, we present a new integral equation approach to solve the EEG forward problem which delivers computationally stable results independently of the conductivity ratio. This is obtained by leveraging on a mixed conforming discretization together with a suitably designed problem splitting scheme. Numerical results corroborate the analysis and demonstrate the effectiveness of the technique.