I’m trying to temporally concatenate a number of subjects’ resting-state scans so I can extract resting-state networks using group ICA (MELODIC). I tried to concatenate them using fslmerge -t but got an error saying I don’t have enough memory. I have 32GB of RAM which apparently isn’t enough for all of the subjects I’m trying to merge. Does anyone have a suggestion on how to temporally concatenate ~120 scans that won’t require so much memory?
MELODIC already solved this problem for you. You don’t have to concatenate your data, just pass a comma-separated list of files as input, and it’ll run group ICA using their MIGP approach for temporal concatenation.
See the docs for details.