Computed basis functions for finite element analysis based on tomographic data.

Area of Science:

• Bioelectromagnetics
• Computational Biology
• Medical Physics

Background:

• Accurate computation of electromagnetic fields is crucial in bioelectromagnetics.
• Voxelized models derived from tomography present computational challenges.
• Existing methods can be computationally intensive for high-resolution bioelectromagnetic models.

Purpose of the Study:

• To introduce a novel finite element method (FEM) for bioelectromagnetic simulations.
• To address the computational demands of fine voxel grids in bioelectromagnetic modeling.
• To improve the efficiency of electromagnetic field computations in complex biological tissues.

Main Methods:

• Developed a FEM using a regular mesh of cube elements, each containing multiple voxels.
• Implemented element basis functions that approximate interface conditions between different tissue types within an element.
• Validated the method using a test model of nested dielectric cubes and a realistic brain model derived from MRI data.

Main Results:

• The novel FEM showed good agreement with conventional FEM for electrostatic potential in a dielectric cube model (1.5% rms difference).
• Simulations on a detailed brain model yielded potentials differing by only 1% from voxel-based FEM.
• The new method significantly reduced the global finite element matrix dimension by over 50 times compared to voxel-element approaches.

Conclusions:

• The proposed finite element method offers an efficient and accurate solution for bioelectromagnetic field computations.
• This method effectively handles complex, multi-tissue voxelized models common in bioelectromagnetics.
• The significant reduction in matrix size suggests improved computational feasibility for high-resolution bioelectromagnetic simulations.