Hi Alissa,
I can think of a few possibilities for getting rough intracranial
volumes, hopefully someone else will reply if I am either wrong, or
missing something better...
In rough order of descending accuracy and increasing ease:
- manually trace dura/scalp/whatever on slices, using e.g. *MRIcro*,
or maybe try a specialist segmentation program such as *itksnap*
- use something like *FSL BET* or *BrainSuite* to get a skull/scalp
segmentation, fill this (maybe requires convex hull and hole-filling
operations, with e.g. Matlab im proc toolbox, or I think, something
like *mipav*
- add together native (or modulated-warped possibly) segmentations for
GM+WM+CSF, maybe threshold (e.g. > 0.1), then sum over all voxels
- use a tape-measure to get head circumference (!) -- may not be as
bad as it sounds... I'd guess the correlation between this and
intracranial volume is fairly strong, though I can't cite any papers
for that ;-)
Best,
Ged.
P.S. Googling any phrases above *like this* should get the relevant
pages at the top.
Alissa Arc wrote:
> Hello SPM members,
>
> I am conducting a VBM study using SPM5 and would like to specify
> intercranial volumes as a covariate. How would I get the values for
> intercranial volumes?
>
> Alissa
>
|