A Finite-Difference Solution for the EEG Forward Problem in Inhomogeneous Anisotropic Media

Area of Science:

• Neuroscience
• Biophysics
• Computational Science

Background:

• Accurate electroencephalography (EEG) source localization relies on detailed head models.
• Head tissue geometry and properties significantly affect neural signal propagation.
• Current methods often lack flexibility in integrating detailed physical properties.

Purpose of the Study:

• Develop a computationally efficient finite difference method (FDM) for head modeling.
• Enable flexible integration of voxel-wise conductivity and anisotropy data.
• Reduce numerical error in FDM simulations.

Main Methods:

• Implemented a novel FDM solution for head modeling.
• Validated numerical accuracy against analytical solutions for spherical models.
• Assessed computational efficiency against alternative modeling approaches.
• Applied the FDM tool to high-resolution magnetic resonance (MR) images.

Main Results:

• The developed FDM solution demonstrated high computational efficiency.
• Achieved low numerical error, comparable to analytical solutions.
• Successfully integrated detailed voxel-wise conductivity and anisotropy information.
• Showcased the tool's application on real subject MR imaging data.

Conclusions:

• The developed FDM provides a more precise head modeling approach.
• Enhances the reliability of EEG as a brain imaging tool.
• Highlights the value of incorporating detailed tissue properties into head models.