Dear Patricia,
It is not OK to use these results as the BET masking is clearly very bad.
You need to try some other BET options to get better results.
Also, if your images have a lot of neck then it will probably help to crop the
images to begin with (leaving all the brain but getting rid of most of the neck).
If your images have a very different orientation to the standard image
(e.g. more than 45 degrees different) then it might help to run
fslreorient2std on these images also, prior to running sienax.
All the best,
Mark
On 10 May 2012, at 19:22, SUBSCRIBE FSL Patricia Pires wrote:
> Hello everybody!
>
> I'm using sienax to quantify white matter volumes of the subjects in a sample. When I look sienax outputs with the following command:
> sienax Control_coADR_371.nii -d
> and i open the result of the white matter segmentation I_stdmaskbrain_seg_2.nii.gz, in the image I see that there are places that did not take into account the white matter and i believe it is because it uses a smaller mask (image attached: red = white matter, yellow = gray matter, blue = spinal fluid). It also happens if I use the command -B "f -0.3" and -r option. Is it right if I use these results even if there areas in which ignores white matter? or may be is there an option to increase sienax mask?
>
> Thanks in advance!
>
> Patricia
>
> <sienax_wm_mask.png>
|