Miranda,
I'll assume for now that what you mean by 'cluster' is not a cluster in the sense of clusters produced by a cluster-based correction for multiple comparisons, which was only very recently introduced to qdec (actually, just a script generator is produced). But rather, I think you mean the areas of significance that remain after you have either manually adjusted the threshold or just left the thresholds at their defaults. If this is what you mean by cluster-size, there is no way now in qdec to get the size of this 'cluster', outside of drawing a label around it, and then click on the 'Map Label to Subjects' options, and looking for the text that says 'Found 389 points in source label.' (where in this example, the label had 389 points/vertices).
If you want a bit more exact way, you can use tksurfer:
tksurfer fsaverage lh inflated -overlay /path/to/qdec/sig.mgh
to load the sig.mgh file (from your qdec contrast dir of interest), then adjust the thresholds (View->Configure->Overlay), then click on the 'cluster' you want (the blob), then click the 'Custom Fill' icon (a paint bucket), then click 'Up to functional values below threshold', click Fill, and it will fill that 'cluster' and the terminal window will show you the number of vertices filled. I'm hoping to add this 'fill' feature to qdec at some point, to give a label creation option other than drawing.
Nick
On Tue, 2009-01-13 at 16:43 -0600, nels5614@umn.edu wrote:
Hello,
I am wondering if there is an easy way to determine the size of a cluster of significant voxels resulting from a QDEC analysis? I have created a label and mapped the label to all of the subjects, but didn't know if there was a way to extract information about cluster size.
Thanks,
Miranda _______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer