Hi,
First off, sorry about my misinterpretation of how you were converting
from DICOM!
> - Posterior to Anterior is pixel coordinate flipped from Analyze to Nifti
I think this is more likely to be a peculiarity of Analyze than a
mistake in the NIfTI... see e.g. the rather complicated hist.orient
field discussed here:
http://www.grahamwideman.com/gw/brain/analyze/formatdoc.htm
> - Physical coordinates are offset between analyze and nifti
I agree with Mark that the nifti origin looks suspiciously far away
from the image centre. There doesn't seem to be any obvious
relationship between the Analyze origin and the nifti qform's 4th
column...
> - The nifti coordinates show that the Y and the Z axes are coupled
> (i.e. z changes if I change y and vice versa).
> I didn't expect this and if Y is coupled to Z, I would expect X to
> be coupled also (it's not).
This however, isn't necessarily wrong, in my opinion. This coupling
between the Y and Z axes (but not the X axis) simply results from a
small rotation around the X axis. This is also what the qform below
encodes.
> qto_xyz:1 -3.281250 0.000000 -0.000000 638.135620
> qto_xyz:2 0.000000 -3.275130 -0.305243 644.623718
> qto_xyz:3 -0.000000 -0.200316 4.990674 11.303008
> qto_xyz:4 0.000000 0.000000 0.000000 1.000000
It could simply be that the debabeler is taking into account a
rotation specified by the DICOM direction cosines that SPM2's Analyze
conversion ignores (because Analyze wouldn't be able to properly
handle this anyway). Likewise, it seems possible to me that the NIfTI
origin could be correctly set from the (peculiar) origin in the
DICOMs. Although it does look odd...
I guess the best thing to do would be to try some other DICOM->NIfTI
conversion tools and see how they compare. If you're very keen, you
could investigate the DICOMs directly, but it can get rather confusing...
Best,
Ged.
|