Information Technology Reference
In-Depth Information
0.6
0.5
0.4
0.3
0.2
0.1
...
.
¡
0
-2 -1 0 1
2 3 4
5
6 7
-0.1
…
t
1
t
2
t
3
t
6
... .
(a) Input vector
... .
Training Samples
... .
Single-Voxel
Classifier
t
1
t
2
t
3
t
6
Task
type
... .
…
sample
1
t
Training
i
sample
2
.
..
.
sample
3
... .
…
sample
n
... .
Task
type
Test Dataset
…
?
?
?
Data
1
Data
2
Data
Predicting
?
t
j
n
(b) Training single-voxel classifier
Single -Voxel
Classifier
Prediction
1
voxel1
Type
Unknown
Dataset
Classifier
voter
Predicted
Type
Single -Voxel
Classifier
Prediction
n
voxeln
(c) Multi-voxel classifiervoter
Fig. 5.
TheschemeofSVVC
is the input of each single-voxel classifier. The details of the two kinds of classi-
fication are given in the following sub-sections.
Multi-Voxel-Classifier with the Sum of BOLD Signal Change (MVC)
As shown in Figure 4, the input vectors of the MVC consist of the sum of
BOLD changes of voxels in predefined regions, the classifier are trained from
the vectors of training samples along with corresponding labels. Once trained,
the classifier can be used to predict labels for a test set. The predicted labels
are then compared to the true labels and the accuracy of the classifier can be
computed.
