Hi,
The afni2avw script is very old and not something we
would recommend anymore. What this issue sounds
like is a difference in how the coordinate system is
set up and how it is displayed.
Have a look at the output of fslhd on your two files.
Are they the same for dim1-dim3 and pixdim1-pixdim3?
Are they different in qform and/or sform?
I suspect that this is what has happened, and that
MRIcron takes the qform/sform information (where
the coordinate origin is, etc.) into account when displaying
the images, whereas this is not taken into account when
fslview displays images.
So the question is whether this is enough for you or not.
Do you need the images to be aligned in the viewer?
If so, then something will need to be done about the
qform/sform (assuming this is what it is). Note that
flirt has the capacity to use the qform/sform info to
align images - try the -usesqform option together
with the -applyxfm option. Otherwise, just register
the images together in the standard way.
All the best,
Mark
Tobias Teichert wrote:
> Hi everybody,
>
> I am encountering very strange behavior in fslview: I used 3dAFNI2NIFTI to
> convert two afni BRIKs to nifti. One of the volumes is a highres image, the
> other an atlas. If I look at the nifti images with fslview, the atlas does
> not match the highres. The differences are not huge, but clearly noticeable
> (it is not just a left-right issue either). If I look at the same two images
> with MRIcron the two images match perfectly!
>
> Seems to be a conversion problem. Thus, I tried to use a different
> conversion routine and found a reference to fsl's afni2awv routine on the
> net. However, my fsl-installation (4.0) does not seem to incorporate this
> routine. I also checked for similar routines like afni2fsl or afni2nifti,
> but I did not find anything.
>
> Any help is greatly appreciated!
>
> Best,
> Tobias
>
>
|