 |
 |
 |
|
 |
|
 |
|
 |
|
 |
|
 |
|
 |
|
 |
|
 |
|
 |
Home > Feature Story: Influence of Local and Remote White Matter Conductivity Anisotropy for a Thalamic Source on EEG/MEG Field and Return Current Computation
Feature Story:
Influence of Tissue Conductivity Anisotropy on Return
Current Computation in a Realistic Head Model
C.H. Wolters, A. Anwander, X. Tricoche, S. Lew and C.R. Johnson
Surgeons, neurologists, and neuroscientists are often interested in non-invasively localizing the neuronal source of brain activity. The accurate localization of neuronal activity is important for reasons ranging from the surgical treatment of epilepsy to the study of brain function. Electroencephalography (EEG) and magnetoencephalography (MEG) are often used for the non-invasive measurement of brain activity. EEG is a method where conducting electrodes placed on the scalp are used to measure the electrical field generated by an active group of neurons on the brain surface or even deep within the brain. MEG is similar to EEG, except that the magnetic field generated by active neurons is detected. Because a set of EEG or MEG measurements can be the same for multiple neural source configurations (a technical situation known as being ill-posed), using EEG and MEG measurements for the subsequent localization of neuronal activity is often difficult and requires advanced mathematical forward and inverse solution techniques.
Fig. 1. Segmented five tissue head model: skin (blue), skull (light blue), CSF (green), gray matter (yellow) and white matter (red). |
The forward solution takes a specified neural source within the head and projects that source to the sensor array. In the case of EEG this means projecting the electrical field from the neuronal source through the head to the scalp surface. One can think of the forward problem as estimating the effect (voltages on scalp surface) from a particular cause (neural source). And, in the case of MEG this means projecting the magnetic field from the source to the location of the MEG sensors just outside of the head. Inverse methods are the opposite of forward methods, the field from the sensors is used to calculate a source location, or estimating the cause (neural source) from the effect (EEG or MEG measurements). Inverse solutions are complicated by the fact that for a given field detected at the sensors there exist no unique source configuration. For this reason, the inverse problem is ill-posed. The use of forward and inverse solutions is often complicated by the complexity of the conduction of electrical and magnetic fields through the various tissues of the head. Tissue components such as cerebral spinal fluid (CSF), neuron cell bodies (gray matter), neuronal axons (white matter), fat, bone (skull), and skin (scalp) have different conduction properties (see figure 1). Conduction properties are specific to tissue types. For instance, white matter conductance properties are very sensitive to direction or are highly anisotropic. The most advanced mathematical solutions used for neuron source localization rely on accurate measurements of tissue conductivity for both electrical and magnetic waves.
In this feature article, Wolters, et al. examine the influence of white matter conductivity on both forward and inverse solutions. The authors found that for specific EEG and MEG formulations the local presence of electrical anisotropy in the tissue surrounding the source substantially compromised the forward field calculation and the inverse solution. The degree of error resulting from the uncompensated presence of tissue anisotropy depended strongly on the proximity of the anisotropy to the neuronal source (see figure 2).
 Fig.2: Surface return current for the left thalamic source in the isotropic model (left) and in model /aniso_thalaniso /with 1:10 white matter anisotropy (right). |
Publication
C.H. Wolters, A. Anwander, X. Tricoche, S. Lew, C.R. Johnson. “Influence of Local and Remote White Matter Conductivity Anisotropy for a Thalamic Source on EEG/MEG Field and Return Current Computation,” In Proceedings of The Joint Meeting of The 5th International Conference on Bioelectromagnetism and The 5th International Symposium on Noninvasive Functional Source Imaging within the Human Brain and Heart, pp. (accepted). 2005.
