Hi,
You cannot change the effective voxel size in the qform as it
only contains information about rotations and translations.
What you are talking about is a change in the basic voxel size.
This has to be done via the pixdim fields, which can be achieved
via the fslchpixdim command. If you want to retain your previous
sform and qform information too (modulo change in voxel size
for the qform) then you'll need to take a copy of that beforehand
and reset it afterwards - both of which can be done with fslorient.
Anyway, that's the FSL solution. If you've already found another
way then that's fine, and neurological should work fine, although
be warned about using it with diffusion data as it can cause some
problems still.
All the best,
Mark
On 28 Oct 2009, at 16:31, Greg wrote:
> I got around this problem using afni (to3d) to convert the dicom
> images to
> nifti (forcing the slice thickness to be equal to the i-s gap).
> Previously I
> was using dcm2nii. Of course now FSL reads the images as
> neurological, but
> that is an easier fix. I understand why the image converters are
> interpreting the images as 2.4 mm thick, I just don't know why
> fslorient and
> setqform won't allow me to reset the thickness to the proper value
> in the qform.
>
|