Hi,
It is possible to use BET to calculate the brain volume but not
the intercranial volume. It is also not the most accurate way of
measuring the brain volume either.
Instead the best way to measure the ICV is by using SIENAX
to derive a normalising factor which is based on the skull (and
hence ICV). This factor can then be used to normalise your
FIRST volumes (multiply the volumes by the VSCALING value
in the sienax report to get the normalised volumes).
I would recommend reslicing to axial and cropping with FSLROI
before running SIENAX. You will also need to *carefully* examine
the report from SIENAX to make sure that the BET has worked
well. I found that using the -R and -f 0.3 worked on the E05
image, but it may vary for your different images. The command
I used was:
sienax E05_cropped.nii.gz -o E05_sienax -B "-f 0.3 -R"
Do make sure that you try different BET options in the -B part
and check that the results are good before accepting the output.
Once you are happy just look in the output directory for the
report.sienax file and inside this you will find the VSCALING
value.
All the best,
Mark
P.S. To answer your question about BET and neck - I recommend
cropping the original images with FSLROI to remove the neck.
Alternatively, you can use the -c or -R option with BET, although
there are other advantages to doing the cropping.
On 9 Nov 2009, at 08:55, SUBSCRIBE FSL hankee wrote:
> Hi, I want to get subjects intracranial volume. So I search some
> related
> message in this archive and get some procedure about it. Procedure
> just likes
> below:
> 1. using bet command: bet E03.nii.gz E03_brain.nii.gz -f 0.4 -g 0
> 2.avwstats E03_brain.nii.gz -V
>
> At step 2, the error showed up and said no such command (I use FSL
> 4.1.4
> version). So I don't know how to get the intracranial volume in this
> way?
>
> By the way,
> 1.when I use fslview tool to see E03_brain.nii.gz file, I find some
> neck tissue
> still in image. If the avwstats command can work, those tissues (eg:
> neck)
> volume will include in intracranial volume result?
> 2. How to edit bet command to reject neck tissue completely?
>
> Thank you very much!!!
>
> yours sincerely:
> Hankee
>
|