Hi - yes, in theory FAST should work fine - that's effectively what the
SIENAX script is doing - running BET and then FAST, to measure things like
atrophy.
By default, FAST doesn't use templates (eg priors), so in general you
don't need to worry about the second question.
Indeed - most lesions in T1 get classified as grey; the darker lesions can
get classified as CSF. Even if you increase the number of classes that
FAST is trying to find, you probably won't have much joy - they just DO
look like grey. (Even turning on priors option (-A) isn't much use as the
prior images that everyone uses are so blurred.) To get more success, you
also probably need either T2 or FLAIR as well, and either feed the pair
into mfast, or something more cunning; we are working on a simple script
to combine two-modality FAST outputs, but it's a messy problem.
thanks, Steve.
On Fri, 14 Feb 2003, Jay Ives wrote:
> Hi,
>
> Can FAST be relaibly used in patients with atrophic brains? Would an
> appropriate set of templates improve accuracy?
>
> I have found that segmentation of 3D T1 weighted volumes classifies white
> matter lesions as gray matter. How can I overcome this?
>
> Thanks.
>
> J
>
Stephen M. Smith MA DPhil CEng MIEE
Associate Director, FMRIB and Analysis Research Coordinator
Oxford University Centre for Functional MRI of the Brain
John Radcliffe Hospital, Headington, Oxford OX3 9DU, UK
+44 (0) 1865 222726 (fax 222717)
[log in to unmask] http://www.fmrib.ox.ac.uk/~steve
|