Low-frequency Functional Connectivity in fMRI - Biomedical Signal Analysis: Contemporary Methods and Applications - page 272

TMP pred TMP corr MFE SOM

FVQ

FSM cc-cluster

Figure 9.5
Segmentation results in the motor areas of subject #1 in the resting-state. The
obtained functional connectivity maps are shown for all unsupervised methods. The
cc-cluster describes a method based on the threshold segmentation of the
pixel-specific correlation map of the motor stimulation fMRI experiment. This map
assigns to each pixel the Pearson correlation coecient between the PTC and the
time-delayed stimulus function. The threshold was chosen as Δ = 0 . 6, and thus
every pixel correlation exceeding 0.6 is considered as activated and is white on the
cc -cluster map. The procedure used in order to obtain the segmentation for
clustering of the resting-state data is explained in the text. The average value of all
PTCs belonging to segmented areas determines a segmentation representative PTC
shown under the respective assignment map.

method in both variants yields acceptable results compared to the other