Dear FSL experts
Sorry to post again but real need answer to this one
Can I ask a question
I am running an analysis of patients v controls
I would like to assess changes in DMN between the groups
Have run:
1. Group ICA (all data sets)
2. Dual regression
I am now trying to use randomise to look for significant changes - unpaired t-test
Other groups seem to have been able to show significant changes by constraining randomise to a pre-defined mask of say DMN or Visual taken from the group maps.
Is this ok? I am only interested in individual network changes rather than whole brain and am worried will lose power by using the whole brain mask in randomise
Thanks!
Monty
|