hi Charalampos
As Alexander says, probabilistic atlases are always going to give you
more accurate results than atlases based on a single brain. If you
are looking up points in the Talairach Daemon, the region labels were
derived from a tracing of the 1988 Talairach Atlas...thus, a single
brain (and probably not the most representative single brain, at
that). You may want to consider looking into the probabilistic
atlases mentioned earlier on this thread for some alternatives.
Another point to keep in mind is that looking up peak coordinates on
ANY atlas may be misleading, with regard to the localization of the
peak, the extent of a cluster, or both. So, it's always worth
spending some time with your data and a good (printed) atlas, to get a
good sense of the anatomical localization. (When possible, looking at
single subject data can also be a very helpful thing to do.)
For a helpful and more in-depth discussion, you might be interested in:
Devlin JT, Poldrack RA (2007) In praise of tedius anatomy. NeuroImage
37, 1033–1041. http://dx.doi.org/10.1016/j.neuroimage.2006.09.055
Hope this helps,
Jonathan
On Thu, Jan 28, 2010 at 12:19 PM, Charalampos Styliadis
<[log in to unmask]> wrote:
> hallo ...since i am interested in what was already mentioned let me drop a line
> also...i conducted a SAM group analysis ...i have normalized the MRI and the
> volumetric images in SPM for all subject and run statistics using SnPM... after
> getting the MNI coordinates from SnPM i could not find a way to label these
> coordinate using SPM so i did the following..i used ginger ale to convert the mni to
> talairach coordinates and then used talairach client in order to get a labelling for
> the coordinates...since it is a group analysis and having read what you already
> said do you suggest a better and more accurate way to get the labelling??
> thank you very much
>
> haris
>
|