Multimodality Techniques

A growing body of evidence suggests that there is a good if imperfect correspondence between neural electrical activation and the fMRI BOLD response and that convergent information can be used to improve the reliability of macroscopic electrophysio-logical techniques. Because of the ambiguity associated with the neural electromagnetic inverse problem, a number of investigators have pursued the strategy of using fMRI to define the locations of activation while using MEG or EEG to estimate time courses. Although this approach may have considerable value, it also has its pitfalls. In general, there is no guarantee that activation seen in one modality will be apparent in the other. The relationship between the precise areas of increased bloodflow and electrophysiological activation is not certain. If we assume that anatomical MRI constrains the location and orientations of possible source currents and that fMRI provides an estimate of the identity and relative strengths of active voxels, it is possible to compute the field topography associated with an extended source of arbitrary shape and size. Alternatively, fMRI can be employed as a method to seed dipole source estimates, which are optimized subsequently by using standard nonlinear procedures. This strategy provides a measure of flexibility to account for mismatches between assumptions and source model.

Other investigators have employed a form of weighted minimum norm to combine fMRI and NEM data. The inverse solution is constrained to lie within the cortical surface, and source current orientation may be constrained to lie normal to the local surface. By weighting the reconstruction according to the spatial pattern of apparent activation disclosed by fMRI, it is possible to guide the minimum norm reconstruction to preferentially place current in those regions. Because Bayesian methods explicitly employ prior knowledge to help solve the inverse problem, they provide a natural and formal method to integrate multiple forms of image data. The simplest strategy is to use fMRI data as a prior. This method can profitably employ strategies for the analysis of fMRI data that quantify the probability of activation in any particular voxel on the basis of fMRI data. Bayesian methods will also benefit from the efforts to develop probabilistic databases of functional organization. Figure 8 illustrates two approaches employed for the integration of MEG and fMRI data.

Anxiety and Depression 101

Anxiety and Depression 101

Everything you ever wanted to know about. We have been discussing depression and anxiety and how different information that is out on the market only seems to target one particular cure for these two common conditions that seem to walk hand in hand.

Get My Free Ebook


Post a comment