Hello,

so i have a bunch of data sets that i need to subdivide into anatomical areas in preparation for a DTI analysis of connectivity between the subdivided regions. to do this, it was suggested that i use this flirt command :

flirt - in mask_with_anatomical_regions -ref mri_brain_scan -out mask_warped_to_brainscan_space

which makes sense to me, i'm warping the 116 regions in the TT_N27 brain to the same space that my real brain occupies, and then i can divide up the real brain using those regions and send it off to the DTI dpt. its kind of hard to describe the problem so i'll just post up some screens of what flirt does:



so it does that when i'm using my real brain as a reference, and the regions of interest as the input.

however, a strange thing happens when i do it the opposite way, ie, i use the TT_N27 as the reference and the real brain as as the input - at first it gave the same weird stuff you see above, but when i used 3drefit to rotate the real brain (lets call it mybrain) after some trial and error i got it to map to the TT_N27 space if i used flirt on mybrain when it was upside down and backwards. however, i can't seem to get this to work the other way around (mapping the ROI's from TT_N27) to mybrain's space.

if anyone has any ideas or suggestions i'd be most grateful, thanks. all i really want to do is get the ROIs mapped to mybrain's space and i don't really care how its done - although i'd like them to match as closely as possible. i'm supposed to use fnirt eventually to tidy up the ROI's once they're mapped to mybrain's space but i guess i'll cross that bridge when i get there.

sorry if this is a dumb question, i'm new to afni.

thanks

Russell