Ah - it is the --interp=nn that is the problem.
Just get rid of that and you'll see output which is not
binary and then you can re-threshold it as is most
appropriate.
I'd avoid using nearest neighbour interpolation
in general. Much better to get the floating point
output and re-threshold in my opinion.
All the best,
Mark
On 27 Oct 2009, at 15:15, Anastasia A Ford wrote:
> Mark,
>
> Thank you for your suggestion. I tried using the -d float option,
> but when
> my mask is registered to the original space is it already a binary,
> so I am
> not able to look at the values and re-threshold the mask myself. I
> used the
> following command to register back to the original space:
>
> applywarp --ref=../dti_FA.nii.gz --in=L_CP_FMRIB58.nii.gz -d float
> --warp=FMRIB58_2_FA_inweight_stroke_nonlinear_transf.nii.gz
> --out=L_CP_native_Oct27.nii.gz --interp=nn
>
>
> Thank you,
>
> Anastasia
>
|