Dear FSL expert,
I'm running randomise on a 4D file containing the default mode network (DMN) resting state information for 19 subjects. The command line I am using is:
randomise -i dr_stage2_ic0012 -o Randomise/dr_stage2_ic0012 -d Randomise/Design.mat -t Randomise/Design.con -f Randomise/Design.fts -n 5000 -T
My design consists of 19 subjects, 3 groups (7, 5, 7). I have 9 t-contrasts and 1 f-contrast.
When I look at my results I only get activation in one small area of the (mid)brain (not part of the DMN). The results give the impression that I used a mask file to limit my search to one region yet this is not the case. In addition, I've run this analysis before (same set of subjects) and then I did get normal DMN differences between the groups.
Has anyone come across something like this before? Is there something wrong with how I am running randomise?
(my initial 4D file looks fine when I look at it in fslview, i.e. it has a DMN for each subject/volume)
Thanks in advance!
Rozanna Meijboom
|