open-discussion > Seed-Based analysis using individual ROIs
Showing 1-3 of 3 posts
Results per page:
Apr 19, 2016  01:04 PM | Pablo Grassi
Seed-Based analysis using individual ROIs
Hi Martin,

Nice toolbox! I was wondering if there is an easy way to do a seed-based analysis using individual functionally defined ROIs (i.e. different ROIs for different subjects). This is apparently not implemented yet, right?

Pablo R. Grassi
Apr 20, 2016  09:04 AM | Martin Göttlich
RE: Seed-Based analysis using individual ROIs
Hi Pablo,

if I understood you correctly you want to perform a seed-based analyses in native space. This is in fact not supported. I'll keep your request in mind for future versions.


Apr 20, 2016  02:04 PM | Pablo Grassi
RE: Seed-Based analysis using individual ROIs
Hi Martin,

Thanks for the quick answer! I would indeed prefer to perform the analysis in native space and then transform everything into MNI-space (for the group analysis). But this is not really what I meant; sorry for that.

I have different functionally defined ROIs (e.g. LOC and MT) for all of my subjects = every subject has his own individual MNI-ROIs. What I would like to do is to perform a beta-series correlation using these functionally defined individual ROIs (all in MNI-space). Thus, in the analysis definition (i.e. anadef.m) I would need to have individual ROIs (and not only a one group ROIDir and ROINames).

Probably the easiest way to do this is to change the variable AnaDef.ROIDir to a subject based structure (e.g. AnaDef.Subj{subj}.ROIDir). Right? Then the important thing would be to modify the BASCO.m file and transform the ROIFile and ROINum variables to subject dependent variables. This would allow one to use individually defined ROIs for the different analyses (and would only require to modify the BASCO.m file by adding a couple of for-loops over subjects).

It would be cool to add this option to the toolbox! (also, since its a common practice to use individual ROIs). Please consider it for future versions!