> I always assumed that the T1 template in spm was derived (or even
> identical) to the ICBM standard of a bunch of averaged T1s "ICBM 452 T1
> Atlas" or previous versions, downloadable at
> http://www.loni.ucla.edu/ICBM/ICBM_ICBMAtlases.html
The MNI ICBM data and the UCLA ICBM data are actually in quite different
spaces. I'm not sure which of these is likely to become the ICBM standard.
>
> At this website, some nice probability maps of brain regions and nuclei are
> available, we want to use them as ROIs for a few studies (with data
> registered to MNI). Unfortunately, I noticed after downloading that these
> images are NOT in MNI space, at least the origin is not at the AC. It is
> set at [0 0 0]. I found some references to this in the list:
> http://www.jiscmail.ac.uk/cgi-bin/wa.exe?A2=ind0311&L=spm&P=R59315&I=-1
> http://www.jiscmail.ac.uk/cgi-bin/wa.exe?A2=ind0311&L=spm&P=R57124&I=-1
>
> So apparently MNI ~= ICBM ? (in Matlab, MNI !=ICBM for C programmers).
> Unfortunately I didn't find much on how to solve this.
I'm afraid that I don't have any bright ideas either. I hope that there will
be some kind of resolution in the next ICBM meeting, preferably with some
sort of policy statement.
>
> Does anyone now if, and perhaps how, I can transform the images from ICBM
> to MNI? All the images downloadable at ICBM seem to occupy the same space,
> so I guess if I can transform one to MNI I can transform them all. I could
> of course affine coregister the ICBM template and ROIs to the MNI template
> in SPM2 (and I will if there is no other solution), but if this is a known
> and fixed transformation I'd prefer to use that.
If you find anything out, then please tell me.
Best regards,
-John
|