Hi there,
thanks for all the feedback and similar mailing list threads I got regarding my melodic issues. Just to make sure that I am doing it right, maybe someone could comment on the following.
I apply melodic to the preprocessed (feat) and in standard space fmri data on all subjects no matter the group. This is done without specifying the design. Then I run dual_reg with the design files. In the latter I'd also include confound variables, ie I'd include a 3rd EV (1st EV is control, 2nd is patient) marking problematic scans with 1, the rest stays 0. No need to include motion parameters or running fsl_regfilt to denoise?
Thanks very much in advance, your help is really appreciated!
Torsten
|