Dear Alfonso and CONN community,

I am having some trouble getting my subject-space cerebellar lobules created with the SUIT atlas (and transformed from SUIT space back to native space) to show up in the Analyses (1st-level) tab - the ROI signals are flat for all lobules across all participants.

In Setup.ROIs, I am marking the Atlas File box as well as Subject-Specific ROI box. The Setup tab was able to read all 34 lobules within the file.

My guess is that the cerebellum was not in the brainmask used for analysis.
In the 'options' tab, I changed the Analysis space dropdown menu to 'Surface: same as template (fsaverage)'. I oversaw this mistake originally. I also said 'Implicit mask (subject-specific)' for the Analysis mask.
I then pressed Done, overwrote all the files, then Denoised and overwrote, then the First-level analysis and overwrote.

Still no signal showing up for cerebellar ROIs in the 'Source timeseries' window in Analyses (1st-level).

What do you think I could try next? Running out of ideas.
Should I start the project from scratch?

Thank you so much!

All the best,

Fred

Dear Fred,

The brainmask does not apply to ROIs so my guess is that the problem may be in the coregistration between your ROIs and the functional data. I would suggest to try going to the Setup.ROIs tab, selecting there your ROI file and using the option 'ROI tools. Slice viewer with functional overlay (QA_REG)' or the option 'ROI tools. Display ROI/functional coregistration (SPM)' to see whether your ROI files and functional files are properly coregistered (e.g. if they are not you could try simply changing there the option that reads "from Secondary dataset #1..." to point to a different functional dataset that may be in the same space / frame of reference as your ROI files)

Hope this helps
Alfonso
Originally posted by fdoleireuqui:

Originally posted by Alfonso Nieto-Castanon:

Dear Alfonso,

My ROIs (cerebellum from SUIT, and thalamus from Freesurfer) appear very well coregistered to my T1 scans.

How should we proceed. I'm running my analysis using subject-based individual-space analyses. All ROIs are in subject space.

Thank you so much!

Best,

Fred


Originally posted by fdoleireuqui:

Dear Fred,

I would perhaps suggest to also check whether the functional and anatomical data appears to be properly coregistered (e.g. in Setup.functionals select the option 'Display slice viewer with anatomical overlay (QA_REG)" or the option |Display functional/anatomical coregistration (SPM)" for a few different subjects to get a sense of whether that looks fine as well)

Best
Alfonso
Originally posted by fdoleireuqui: