Hi Simon and Marko,
If you plan to create custom tissue probability maps, there are two
threads that may help you to decide whether you want to smooth it or not.
http://www.jiscmail.ac.uk/cgi-bin/wa.exe?A2=ind0602&L=SPM&P=R56961&I=-3&X=35FF8072475B138179&Y=hedok.lee%40mssm.edu
http://www.jiscmail.ac.uk/cgi-bin/webadmin?A2=ind05&L=spm&O=D&X=35FF8072475B138179&Y=hedok.lee%40mssm.edu&P=304762
Hope this helps,
Hedok
Marko Wilke wrote:
> Hi Simon, All,
>
>> The normalisation and segmentation of images in SPM is obviously
>> based on an
>> asymmetric template for standard VBM analyses.
>
> Yes (as the average brain is asymetrical, so is the average template :)
>
>> However, how do I specify or
>> incorporate symmetric priors for image normalisation in SPM5 for VBM
>> reflection
>> asymmetry analysis? Do I need to create one myself?
>
> To specify them is easy (in the segmentation job, click on custom and
> specify whatever you like as priors), but creating them will be more
> tricky as opposed to the earlier priors, they are not smoothed anymore
> (see grey.nii in spm5/tpm, which is the standard GM prior now). By
> flipping them to make them symmetrical, you will introduce a
> smoothness that was not there before.
>
> You could use the asymmetrical priors to write out native-space images
> and then still use a (symetrical-ized and smoothed) GM partition for
> normalization, but I am not sure this is the optimal way.
>
> Best,
> Marko
>
|