Hi Jan
t2_lesion is just a short shell script wrapper for fast, it will call
bet with the -f 0.3 option and then calls
fast -t2 -e -ov on the bet output. The first option tells fast that
it's dealing with a T2, the second switches on partial volume
classification and the third generates partial volume images as output.
In T2s, lesions and CSF have similar high intensities, so t2_lesion
uses a 3 class fast segmentation to find the lesion+CSF mask.
hope this helps
christian
On Wednesday, Jun 25, 2003, at 09:00 Europe/London, Jan Fenker wrote:
> hello, first i want to thank you for your software which i mainly use
> for
> structural images of patients mit ms. in your bin-library i have found
> the
> script "t2_lesions" which seems to give a better estimation of the
> brain
> percentage in patients with a great lesion load than sienax in t2
> weighted
> images. but i do not really understand where that difference come
> from, as
> far i as understood both use the fast-script but with other options
> and i do
> not find an explanation for the options used in t2_lesion script which
> makes
> it clear to me why it better works in such cases. as i would like to
> understand what i am doing i would be very thankful if you could give
> me a
> hint. and once again congratulation to you for your inventions.
>
> jan fenker
> department of neurology II
> university of magdeburg
|