Hi Paul,
you probably wanted to post this to the SPM list...
From what I have observed, it doesn't really make a difference whether you
normalise your DWI images or the FA image. Local rotations introduced by
nonlinear normalisation do not matter for FA calculation.
The problem is indeed to get a good set of normalisation parameters...
Volkmar
On Wed, 20 Jul 2005, Paul Macey wrote:
> If you plan to normalize before you calculate the tensors, you'll need to
> transform the direction gradients, which is not a trivial task. I would
> suggest calculating the tensors in native space and normalizing the FA maps,
> perhaps using parameters from b0-to-T2 spatial normalization (although
> personally I have found this unreliable - check the accuracy of the spatial
> normalization carefully).
>
> Cheers,
> Paul
>
> >On Wed, 20 Jul 2005, Anusha Sritharan wrote:
> >
> >> I have DTI data for two groups; and I wish to compare the FA differences
> >> in WM between groups.
> >>
> >> I am planning to do the following:
> >>
> >> 1. After normalising the bo and the DWI images, the FA map will be
> >> generated for both groups.
> >> 2. In order to compare FA maps between groups with in WM I ahave copmuted
> >> a product FA.* WM (WM is segmented from high res anatomical and is in same
> >> space as the FA maps).
> >>
> >> 3. FA.*WM will be smoothed.
> >>
> >>
> >> 4. The smoothed product will then be used to do two sample t-test under
> >> basic models in SPM2.
> >
>
>
--
Volkmar Glauche
-
Department of Neurology [log in to unmask]
Universitaetsklinikum Freiburg Phone 49(0)761-270-5331
Breisacher Str. 64 Fax 49(0)761-270-5416
79106 Freiburg
|