Hi,
I'm not sure what you mean by them being oriented sideways.
If the axes are labeled correctly with FSLView then things are
fine. Note that FSLView does not currently have the option
to reorient the display - it always shows the image in the
same way that it was acquired and just labels things accordingly.
By default we do not use the qform or sform information when
doing registration, as we've found it to be a bit unreliable from
certain conversion software in the past. However, you can
turn it on at the command line using -usesqform. Also, if the
images are in quite different initial orientations (more than
90 degrees different) then you can turn on the full search
space (from the Advanced tab in the GUI or with -searchrx -180 80
-searchry -180 180 -searchrz -180 180 on the command line)
as an alternative, if the -usesqform doesn't work well.
So try this and hopefully it will fix the problem.
If not then let us know.
All the best,
Mark
On 28 Mar 2009, at 02:16, Gregg DiGirolamo wrote:
> Hi All, Sorry for what I'm sure is a really simple fix that I'm just
> to new at to know. We have a 3d T1 anatomical in which the slices
> are acquired axially. This means in the NIFTI header the info is:
> qform_xorient Anterior-to-Posterior qform_yorient Inferior-to-
> Superior qform_zorient Left-to-Right The image is correctly labelled
> when seen in fslview, but the coronal and transverse slices are
> oriented sideways when viewed. And more importantly, when flirt
> tries to register the images, the coronal images are being matched
> to the sagital standard and the transverse to the coronal, etc. How
> can we get fsl change the information to that it is presented
> properly and registers the same image orientations with the same
> standard orientations? Thanks for the help. I've moved from SPM to
> fsl and I'm luv'in it! Gregg
|