Print

Print
A couple of things....

Segmentation in SPM12 generally works much better than either of the options in SPM8 (see previous posts) and is likely to give much greater sensitivity in VBM analysis.

I'm not sure if this is the case here, but if you are working with a population of subjects from east Asia, you may want to try changing the regularisation settings for the initial affine registration part of the segmentation. Generally, when the segmentation goes badly wrong, the reason is to do with how the TPMs are initially aligned with the data. By changing the regularisation so that it is more appropriate for your population, things can work much better.

Best regards,
John


Box Lee <[log in to unmask]> wrote:
Hi John and other dartel user,

Recently I come across a odd problem when I am doing DARTEL-VBM analysis of T1W data.
I choose the "New segment" tool to segment tissues of my data. 
But the results of GM volume contain a lot of non brain tissue  including eyes, part of skull and some tissue of neck as the attached images show.(about 1 third subject of my data got this problem). Will these residury tissue affect the results of the following DARTEL processing( create template and normalization)  within the brain. or can I just mask the dartel input  rc1*.nii with a "intra-brain mask".
And I attach some screen shot of my data. 
Thanks a lot.

-
Box Lee