Hi,
Are you trying to look at the coefficient warp images?
These are not strictly nifti images and there is little
point looking at them. It should be possible to look
at the resampled images, the jacobians or the warp
field images without any problem as all of these are in
standard nifti format.
You can convert coefficient files to warp field files or
Jacobians using fnirtfileutils.
If it is not a coefficient file that you are trying looking at,
can you please send us the output of fslhd on the offending
file?
All the best,
Mark
On 18 Mar 2010, at 19:14, John Ollinger wrote:
> I have been trying to use fnirt for spatial normalization in an AFNI
> analysis, and seem to have run up against an inconsistency in the
> header.
> The sform_code and qform_code are both set to 4 in the header, which I
> understand to mean that they both represent the images position in MNI
> space. Unfortunately, the qform indicates that the image
> orientation is RPI
> while the sform indicates that it is LAI. The sform is correct. This
> confuses AFNI to no end.
>
> My understanding of the nifti standard is that if the sform_code and
> qform_code are the same, then the sform and qform should represent
> the same
> coordinate transformation. Am I missing something here?
>
> Thanks,
> John
>
> p.s. This occurs with version 4.1.5.
>
|