Hi Anastasia,
first, if the resolution of your native FA images is different from the FMRIB58, you should not expect the same number of voxels. Second, you have to bear in mind that TBSS is doing a registration of this FA images onto the template that may enlarge or "shrink" a brain and some structures of the brain to make it fit the template the best possible. So you should not be surprised if the number of voxels in your native space is really different from the standard space, and if it differs from one subject to an other.
Cheers,
Gwenaelle
> Hello,
>
> I have a few questions regarding tbss_deproject function. I
> understand that
> statistical files generated by tbss scripts can be
> projected back from the
> skeleton space to original subject's FA map. I am a
> little bit puzzled by
> the outcome of the de-projection. I was expecting that each
> subject's FA map
> will have equal number of voxels per region as that of
> other subjects. Say
> in we have 100 voxels in the corpus callosum that show
> statistically
> significant differences on the skeleton. The these 100
> voxels will be
> de-projected back to the original FA map for each subject.
> Is that correct?
> Or does the percent overlap between the skeleton voxel and
> original FA voxel
> play a role and thus some subjects will have 99, other 80
> voxels?
>
> Sincerely,
>
> Anastasia
_____________________________________________________________________________
Envoyez avec Yahoo! Mail. Une boite mail plus intelligente http://mail.yahoo.fr
|