PDE-constrained Optimization for Electroencephalographic Source Reconstruction

• Published in: Lobachevskii journal of mathematics 2024, Vol.45 (6), p.2875-2894
• Main Authors: Malovichko, M. S., Yavich, N. B., Razorenova, A. M., Golubev, V. I., Koshev, N. A.
• Format: Article
• Language: English
• Subjects: Algebra; Algorithms; Analysis; Differential equations; Discretization; Electroencephalography; Geometry; Inverse problems; Linear equations; Mathematical Logic and Foundations; Mathematics; Mathematics and Statistics; Numerical methods; Optimal control; Optimization; Poisson equation; Probability Theory and Stochastic Processes
• ISSN: 1995-0802; 1818-9962
• DOI: 10.1134/S1995080224603266

This paper introduces a novel numerical method for the inverse problem of electroencephalography (EEG). We pose the inverse EEG problem as an optimal control (OC) problem for Poisson’s equation. The optimality conditions lead to a variational system of differential equations. It is discretized directly in finite-element spaces leading to a system of linear equations with a sparse Karush–Kuhn–Tucker matrix. The method uses finite-element discretization and thus can handle MRI-based meshes of almost arbitrary complexity. It extends the well-known mixed quasi-reversibility method (mQRM) in that pointwise noisy data explicitly appear in the formulation making unnecessary tedious interpolation of the noisy data from the electrodes to the scalp surface. The resulting algebraic problem differs considerably from that obtained in the mixed quasi-reversibility, but only slightly larger. The algorithm does not require the formation of the lead-field matrix, which can be beneficial for large matrices. Our tests, both with spherical and MRI-based meshes, demonstrates that the method accurately reconstructs cortical activity.