Hi there,
> Is there some ways (tests) to make a direct comparison between
> different brain MRI images?
this is not a question very specific to SPM, but I will try an answer
anyway:
> With the MRI images that I have, I managed to create my own templates
> (GM, WM and LCR) by normalizing and segmenting the MRI images of kids
> between the age of 10 to 15 years using VBM5 and now I'm in need of
> some kind of test to prove, for example, that a brain of a 5 year old
> doesn't belong in the class of 10 to 15 years, a kind of test
> that shows a big differences between the voxels of selected region or
> shows the deviation when I compare the mean brain (standard) to the
> one of the 5 year old.
I guess that would depend on the kind of question: you could simply look
at the effect of age on every voxel by setting up a statistical model in
which you use age (and possibly its quadratic and cubic extensions, and
other variables you are interested in) as a covariate. This is the
approach we recently took (Wilke et al., NeuroImage 2008) and we found,
not suprisingly, that age is dominant in explaining the variance in
structural brain changes in children aged 6-18. However, you will need a
certain number of kids in order for this approach to be meaningful. As
this data is publicly available, you could analyze the results with
regard to where the greatest effects of age occurr. You can also take an
indirect approach by normalizing each child to an adult template and see
where the deformations correlate with age, but that is only two sides of
one coin, with the former likely having a higher resolution. To divide
the whole population into different age ranges ("classes") based on
unbiased criteria, that would need some more thinking.
Hope this helps,
Marko
--
=====================================================================
Marko Wilke (Dr.med./M.D.)
[log in to unmask]
Universitäts-Kinderklinik University Children's Hospital
Abt. III (Neuropädiatrie) Dept. III (Pediatric neurology)
Hoppe-Seyler-Str. 1, D - 72076 Tübingen
Tel.: (+49) 07071 29-83416 Fax: (+49) 07071 29-5473
=====================================================================
|