Dear Shahrzad,
> I have a question regarding the use of DARTEL template for a VBM study.
>
> If I have 5 groups of subjects which 4 groups (each group has about 14
> subjects) do NOT have normal brains (because of a special disease the brains
> are highly malformed) and just one group is the control group (with about 23
> subjects) and all subjects are very old, do you recommend to make a DARTEL
> template for them or would it be better to use MNI template?
> brains in different groups are rather different as each group is in a
> specific stage of the disease (except the control group).
I would expect that a high-dimensional registration routine like
DARTEL is going to do a better job normalizing these odd brains, so I
think that is a reasonable way to go.
> If yes, then I assume that I have to use all 5 groups to make the template?
> (even all subjects from the control group?)
You should probably use equal numbers from each group so that the
template isn't biased towards one of the groups. This is another
reason not to use a standard MNI template - this will be biased more
towards healthy young adults.
> I am also wondering if using the new segment option would helpful in my
> case?
Probably, as it contains improvements over the old segmentation.
Though you will have to see how it seems to do with your particular
brains. It would also probably be worth trying the segmentation in the
VBM8 toolbox (which as I recall does not relay on tissue probability
maps, which may help with abnormal brains - though maybe others with
actual experience can comment!).
Hope this helps!
Best regards,
Jonathan
--
Dr. Jonathan Peelle
Center for Cognitive Neuroscience and
Department of Neurology
University of Pennsylvania
3 West Gates
3400 Spruce Street
Philadelphia, PA 19104
USA
http://jonathanpeelle.net/
|