Hi Hoang,
Yes, when you have an image with a large amount of pathology, then the
assimptions of clean Gaussian mixtures for the different tissue classes
starts to break down. For this kind of image, it may help to try a
multi-channel segmentation (ie T1 and T2 and FLAIR at the same time), or
to add in a pre-FAST or post-FAST hack to fix specific problems. I assume
that you are running BET before running FAST.
If you would like to put a sample triplet of images on a website we can
take a quick look.
Thanks, Steve.
On Wed, 14 May 2003, Hoang Tran wrote:
> Howdy All,
>
> When segmenting a T1 brain with severe WM disease, i.e. lots of T2
> hyperintensities, using FAST with no priors, I'm getting a lot of
> misclassification of gray and white matter. In particular, a significant
> amount of gray matter is being classified as white. WM signal
> hyperintensities appear grayish in T1 images, so I would expect that I
> should have more misclassification of white as gray as oppose to gray as
> white.
>
> My observations runs counter to my intuition which makes me question what I
> am doing and possibly my usage of FAST. Has anyone else observed this issue?
> If so, is there something obvious that I am doing wrong. For the record, I
> am segmenting T1 images with the default settings for FAST using no priors.
> I also have FLAIRs (T2s with CSF appearing dark) which gives me a reference
> of the severity of WM hyperintensities. T1s that do not have a significant
> amount of WM hyperintensities, i.e. percentage volume of WM signal
> hyperintensities to white matter is roughly less than 10%, segment properly.
> Any help would be much appreciated. Thank you in advance.
>
> Cheers,
>
> Hoang Tran
>
> Neurobehavioral Research, Inc.
> 201 Tamal Vista Blvd.
> Corte Madera, California 94925
> 415.927.6627 (Office)
> 415-924-2903 (Fax)
> [log in to unmask]
>
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
|