Dear experts,
Currently I'm working on resting state data collected from two different sites with the same scanner and same length of session (6 min)
but they were different in terms of repetition time; One is 2s and the other one is 3s.
What I plan to do is as follows.
1) Run concatenating group ICA for two dataset together: identify robust RSN components across datasets.
2) Run dual regression focusing on stage 1 (estimating individual time course estimation within group-level RSN component) and stage 2 (estimating individual spatial map corresponding to group-level RSN component).
3) Do other analysis outside of fsl, melodic and glm with estimated individual time-course and spatial map corresponding to each RSN component at the individual level.
I simply guess based on previous posts regarding this issue (https://www.jiscmail.ac.uk/cgi-bin/webadmin?A2=ind1103&L=fsl&F=&S=&P=31502)
that it would be okay to the procedure described above without resampling or without separated group-level ICA for each dataset, because I'm not going to do group-level paired comparisons and both dataset are the same in length.
Is my understanding correct? Otherwise, should I run these two dataset separately for concat-group ICA and dual-regression?
I would appreciate any practical guidelines to handle these different TR situation.
Many thanks!
Taeho
|