A Case Study of Focal Bayesian EEG Inversion for Whitney Element Source Spaces: Mesh-Based vs. Cartesian Orientations

This paper concentrates on the Bayesian detection of the neuronal current distributions in the electroencephalography (EEG) imaging of the brain activity. In particular, we focus on a hierarchical maximum a posteriori inversion technique applicable when the lead field matrix is constructed via the finite element method. We utilize the linear Whitney (Raviart-Thomas) basis functions as source currents. In the numerical experiments, the accuracy was investigated using two spherical head models. The results obtained suggest that the interpolation of the dipolar source space does not necessarily bring any advantage for FEM based inverse computations. Furthermore, the divergence conforming Whitney-type sources were found to be sufficient for precise and highly focal Bayesian modeling of dipole-like currents.

[1]  Z. Zhang,et al.  A fast method to compute surface potentials generated by dipoles within multilayer anisotropic spheres. , 1995, Physics in medicine and biology.

[2]  Daniela Calvetti,et al.  Introduction to Bayesian Scientific Computing: Ten Lectures on Subjective Computing , 2007 .

[3]  J. Vorwerk,et al.  Electroencephalography (EEG) forward modeling via H(div) finite element sources with focal interpolation , 2016, Physics in medicine and biology.

[4]  E. Somersalo,et al.  Visualization of Magnetoencephalographic Data Using Minimum Current Estimates , 1999, NeuroImage.

[5]  Martin Burger,et al.  Hierarchical Bayesian inference for the EEG inverse problem using realistic FE head models: Depth localization and source separation for focal primary currents , 2012, NeuroImage.

[6]  Bart Vanrumste,et al.  Dipole estimation errors due to differences in modeling anisotropic conductivities in realistic head models for EEG source analysis , 2008, Physics in medicine and biology.

[7]  Anthony O'Hagan,et al.  Kendall's Advanced Theory of Statistics, volume 2B: Bayesian Inference, second edition , 2004 .

[8]  Martin Bauer,et al.  Comparison Study for Whitney (Raviart–Thomas)-Type Source Models in Finite-Element-Method-Based EEG Forward Modeling , 2015, IEEE Transactions on Biomedical Engineering.

[9]  Ionut Danaila,et al.  An Introduction to Scientific Computing , 2007 .

[10]  J. P. Ary,et al.  Location of Sources of Evoked Scalp Potentials: Corrections for Skull and Scalp Thicknesses , 1981, IEEE Transactions on Biomedical Engineering.

[11]  Cees J. Stok,et al.  The influence of model parameters on EEG/MEG single dipole source estimation , 1987, IEEE Transactions on Biomedical Engineering.

[12]  Thomas R. Knösche,et al.  A guideline for head volume conductor modeling in EEG and MEG , 2014, NeuroImage.

[13]  Maureen Clerc,et al.  Handbook of Neural Activity Measurement: EEG and MEG: forward modeling , 2012 .

[14]  Harri Hakula,et al.  Conditionally Gaussian Hypermodels for Cerebral Source Localization , 2008, SIAM J. Imaging Sci..

[15]  Mark Ainsworth,et al.  Hierarchic finite element bases on unstructured tetrahedral meshes , 2003 .