Numerical mathematics for the modeling of a current dipole in EEG source reconstruction using finite element head models

Carsten H. Wolters, Harald Köstler, Christian Möller, Jochen Härdtlein, Lars Grasedyck, and Wolfgang Hackbusch
(Please use for correspondence this email).

Submission date: 10. May. 2007
Pages: 23
published in: SIAM journal on scientific computing, 30 (2007) 1, p. 24-45 
DOI number (of the published article): 10.1137/060659053
with the following different title: Numerical mathematics of the subtraction method for the modeling of a current dipole in EEG source reconstruction using finite element head models
MSC-Numbers: 35Q80, 65N21, 92C50
Keywords and phrases: source reconstruction, EEG, subtraction method
Download preprint: PDF (1294 kB)

Abstract:
In electroencephalography (EEG) inverse source analysis, a mathematical dipole is widely used as the model of the primary current source. The inverse methods are based on solutions to the corresponding forward problem, i.e., the simulation of the electric potential in the head volume conductor for a dipole in the cortex sheet of the human brain. The current dipole introduces a strong singularity on the right-hand side of the governing Poisson-type differential equation that has to be treated specifically when solving the equation towards the electric potential.
In this paper, we give a proof for existence and uniqueness of the weak solution in the function space of zero-mean potential functions, using a subtraction approach. The method divides the total potential into a singularity potential and a correction potential. The singularity potential is due to a mathematical dipole in an infinite region of homogeneous conductivity (the one at the source position). We then state convergence properties of the Finite Element (FE) method for the numerical solution to the correction potential. We validate our approach using high-resolution tetrahedra and regular and geometry-conforming node-shifted hexahedra elements in a three-layer sphere model. Validation has been carried out using sophisticated visualization techniques and statistical metrics for a comparison of the numerical results with analytical series expansion formulas at the surface and within the volume conductor. Finally, we validate the computed potentials of the subtraction method with the results of a direct approach in realistically-shaped FE head volume conductor models.