A systematic study of head tissue inhomogeneity and anisotropy on EEG forward problem computing

Area of Science:

• Biomedical Engineering
• Computational Neuroscience
• Medical Imaging

Background:

• Electroencephalogram (EEG) analysis relies on accurate forward computation models.
• Tissue conductivity variations, particularly inhomogeneity and anisotropy, are known to influence EEG signals.
• Existing models often simplify these complex tissue properties.

Purpose of the Study:

• To develop and apply a stochastic method for analyzing the effects of inhomogeneous anisotropic tissue conductivity on EEG forward computation.
• To quantify the impact of these tissue properties on EEG signals using a human head model.

Main Methods:

• Stochastic finite element method (FEM) incorporating Legendre polynomials, Karhunen-Loeve expansion, and stochastic Galerkin methods.
• Application to an inhomogeneous and anisotropic spherical human head model.
• Use of Volume and Wang's constraints to define anisotropic conductivities for white matter (WM) and skull tissues.

Main Results:

• Individual incorporation of WM and skull anisotropic properties led to average relative errors of 56.5% and 57.5% in EEG, respectively.
• Combined WM and skull properties resulted in a 43.5% average relative error.
• Inhomogeneous scalp tissue introduced a 27% average relative error, while a full inhomogeneous anisotropic model yielded a 45.5% average relative error.

Conclusions:

• The study demonstrates that inhomogeneous anisotropic tissue conductivity has a significant effect on EEG forward computation.
• Accurate modeling of these tissue properties is essential for reliable EEG analysis and interpretation.
• The proposed stochastic method provides a robust framework for investigating complex conductivity effects in biological tissues.