Biomedical Engineering Reference
In-Depth Information
LORETA current waveforms for all voxels and for each 2-second epoch can be
generated for use in CSC calculations, first by combining them into a priori defined
regions of interest (ROIs) to test hypotheses about circuit function and connectivity.
As an example, we examined data recorded from a healthy adult male during the
resting state and during a simple motor task of repeatedly squeezing either the domi-
nant or nondominant hand (methods as in [144]). Using ROIs to examine connectiv-
ity between the motor strip [Brodmann area (BA) 4] and premotor cortex and
supplementary motor cortex (BA 6), we found that the motor task was associated
with a change in CSC in the contralateral hemisphere's pathway (BA 6 to BA 4)
compared with the resting state (Figure 11.3). We interpret this to indicate that con-
nectivity, as reflected in the shared electrical activity at two anatomically linked
brain regions, may vary instantaneously with the demands of task activation. The
use of this method to study individuals with MDD is currently under way in our lab-
oratory; preliminarily, there appear to be differences both in the resting state and in
task-activated CSC values, and the relationship of these measures to symptomatic
and functional treatment response will be examined.
A limitation to the interpretation of coherence is that it cannot easily discrimi-
nate between the direct coupling of activity between two regions, and a value arising
because both regions are connected to a separate, third area that “drives” the signals
in both probed regions. This has been addressed by the construct of partial directed
coherence (PDC) [146-150]. This method calculates a decomposition of the ordi-
nary coherence function into two “directed” coherences: one representing the
feedforward and the other representing the feedback aspects of the interaction
between two structures. Rather than just revealing mutual synchronicity, PDC
describes “whether and how two structures are functionally connected” [151]. A
refinement of CSC could incorporate such advances in coherence analyses to con-
sider Granger causality [152] and the direction of information flow in these
interactions between brain regions.
4
6
8
5
9
7
3/1/2
40
46
19
10
44
39
45
41
43
18
42
47
22
17
11
38
37
21
20
Figure 11.3
Brain regions of interest.
Search WWH ::
Custom Search