Hi,
It sounds like your sform and qform are not set.
In the nifti image there are codes that say whether there
is trustworthy info in the orientation fields (qform and sform).
See the fslhd output and look for sform_code and qform_code.
If these are both zero then it says there is no orientation
information and fslorient won't be able to change it (although
it reports the "default FSL interpretation" of the orientation).
In order to force any change to happen you need to set either the
sform or qform code to 1 to begin with (which you can do with
fslorient).
I suspect that this is your problem, although it is increasingly
uncommon (and bad) if files are written with no legal orientation
info, so I would check your original reconstructions and processing
pipeline to see if this information is just never created or is lost
somewhere along the line. I only recommend forcing orientation
information into the header if you are *absolutely sure* about the
orientation you are forcing.
All the best,
Mark
James E. Dobson wrote:
> Hi, FSLers,
>
> I'm using FSL 4.1.0 and want to set the orientation in the header for
> our NIFTI files. I've changed the L-R orientation with AIR prior to
> conversion to NIFTI (using fslchfiletype). It doesn't appear that the
> "forceneurological" option modifies my headers. What might I be doing
> wrong?
>
> [jed@eros ANATOMY]$ fslorient -getorient mprage1.nii.gz
> RADIOLOGICAL
> [jed@eros ANATOMY]$ fslorient -getqform mprage1.nii.gz
> 1 0 0 0 0 0.9375 0 0 0 0 0.9375 0 0 0 0 1
> [jed@eros ANATOMY]$ fslorient -getsform mprage1.nii.gz
> -1 0 0 0 0 0.9375 0 0 0 0 0.9375 0 0 0 0 1
> [jed@eros ANATOMY]$ fslorient -forceneurological mprage1.nii.gz
>
> [jed@eros ANATOMY]$ fslorient -getorient mprage1.nii.gz
> RADIOLOGICAL
> [jed@eros ANATOMY]$ fslorient -getqform mprage1.nii.gz
> 1 0 0 0 0 0.9375 0 0 0 0 0.9375 0 0 0 0 1
> [jed@eros ANATOMY]$ fslorient -getsform mprage1.nii.gz
> -1 0 0 0 0 0.9375 0 0 0 0 0.9375 0 0 0 0 1
>
|