Thank you very much Niels,
Do we need to demean the global effects before adding them as an additional confound EV in the design?
Hi,
You can do something like the following:
fslmeants -i all_FA_skeletonized -m mean_FA_skeleton_mask
On Mon, Jan 16, 2017 at 11:41 AM, John anderson <[log in to unmask]> wrote:
Hi Anderson,
Thank you very much for your input!
I highly appreciate if you add more information on how I can extract the global effects, so I can include them as an additional confound EV in the design
Cheers,
John
Hi John,
You can factor global effects out by entering them as a additional confound EV in the design.
Regarding quality control before the statistical analysis, the best way around is to inspect the results of the preprocessing after every step for every subject.
All the best,
Anderson
On 16 January 2017 at 01:05, John anderson <[log in to unmask]> wrote:
Dear FSL experts,
I ran TBSS analysis between three groups using randomise 5000 permutations and TFCE method. I followed all the steps mentioned in WIKI
I got at the end of the analysis, one large cluster represent the difference between the groups, as follow:
cluster --in=tbss_tfce_corrp_tstat1.nii.gz --thresh=0.95
Cluster Index Voxels MAX MAX X (vox) MAX Y (vox) MAX Z (vox) COG X (vox) COG Y (vox) COG Z (vox)
1 116346 1 87 90 22 89 104 82.5
How can I deal with the global effect between the groups, how can I approach the data in this case. How can I investigate if this global effect is a real difference between the groups, or its related to some technical issues? I highly appreciate any suggestions or advice?
John
--
Niels Bergsland
Integration Director
Buffalo Neuroimaging Analysis Center
100 High St. Buffalo NY 14203
[log in to unmask]
|