I am really sorry for this. Did you get the attachments?
The original message was :
Dear FSLers
I am new to FSL and am currently analysing 3 independent groups DTI TBSS data (controls, patients A, patients B). I would really appreciate the confirmation of the design (attached) to run randomise including both -t and -f. I would like to test main effect of group membership and then test separate contrasts between two groups.
If the attached design is correct, then the randomise command should be like randomise -i all_FA_skeletonised -o tbss -m mean_FA_skeleton_mask -d fsl.mat -t fsl.con -f fsl.fts -n 5000 --T2 -V ?
The *_tfce_corrp_tstat* files represent the "post-hoc" output corrected for multiple comparisons?
How can I handle with the p values in the fslview menu? Going through some previous comments in the list, I was confused by one-tailed t-test and two-tailed F-test. Does this difference mean that when visualizing the results from tstat file (resulting from the attached design), the min should be changed to , the min max should be changed to 0.975-1 when using the tstat files and to 0.95-1 when using the f- files?
Thank you in advance for your help.
All the best,
Foteini
Foteini Christidi
Clinical Neuropsychologist
Aeginition Hospital, Neurological Department, Medical School of University of Athens, Athens, GR
University Hospital of Zurich, Institute of Neuroradiology, Zurich, Switzerland
|