Biomedical Engineering Reference
In-Depth Information
|
ˈ
|
in Eq. (
7.68
) is shown in Fig.
7.4
d. In these results, imaginary-coherence-based
images ((a) and (b)), clearly detect the two interacting sources but the seed blur
dominates in the magnitude-coherence-based images ((c) and (d)).
We carried out the numerical experiment simulating gamma-band signals with
theta-band envelopes. The time courses of the three sources for the first trial are
shown in Fig.
7.2
b. The time courses had trial-to-trial time jitters generated using
Gaussian random numbers with the same standard deviation of 20 time points.
The voxel time courses were estimated using the narrow-band adaptive beam-
former with a data-covariance tuned to the gamma frequency band, and the coherence
images were computed. The seed was set at the second source location. The coher-
ence images are shown in Fig.
7.5
a. Here, the magnitude, imaginary, and corrected
imaginary coherence images are, respectively, shown in the top, middle, and bottom
panels. In the magnitude coherence image, the seed blur dominates and only the seed
source is detected. However, since the time jitter created a large phase jitter for the
gamma-band signals, neither the imaginary nor the corrected imaginary coherence
image contain meaningful information on the source connectivity.
(a)
(b)
12
10
8
6
12
10
8
6
y (cm)
12
10
8
6
5
0
5
5
0
5
y (cm)
y (cm)
Fig. 7.5
Results of imaging the gamma-band source coherence on the plane
x
0 cm. The seed
was set at the second source location. The coherence images were computed using the gamma-band
voxel time courses. The
top
,
middle
,and
bottom panels
in
a
, respectively, show the magnitude,
imaginary, and corrected imaginary coherence images. The
top
,
middle
,and
bottom panels
in
b
,
respectively, show the magnitude, imaginary, and corrected imaginary envelope coherence images
=
