On Tue, 20 Jul 2004, K. Oishi wrote:
> I am trying to compare FA maps of the control group with that of patients.
> According to the M. Sach et al (Brain 2004, 127, 340-350),
> normalized FA maps were computed on the normalized diffusion-weighted data.
> I would like to know if it is OK to get normalized FA maps by normalyzing
> indivisual (not normalized) FA maps? What are the differences?
The most important issue will be how you are going to determine your
normalisation parameters. Parameter estimation on the FA maps itself is
probably a bad idea, since FA maps will not fit any template image (e.g.
T1, T2, EPI...) very well.
Since FA is invariant with respect to tensor orientation, it should not
matter whether you do normalisation on FA maps or on diffusion weighted
data, provided you have got the right set of parameters.
Hope this helps,
--
Volkmar Glauche
-
Department of Neurology E-Mail [log in to unmask]
UKE Hamburg Phone 49(0)40-42803-5781
Martinistr. 52 Fax 49(0)40-42803-9955
20246 Hamburg
|