Hi,
I agree with Markus.
The problem is almost certainly something to do with the orientation
information
contained in the nifti images being treated differently by the two
pieces of software
you are using. It is the sform and qform information which is
relevant and should
be the same for the input and output images. Note that if both the
qform and sform
codes are zero then the rest of the information is not used (in this
case default
orientations will be used). Displaying the images in fslview is the
easiest way to
get a feeling for how the rest of the FSL tools will treat these images.
If you find that your other software has removed the qform/sform
information
(which you can see with fslhd) then you can restore it with fslcpgeom
which
copies all the orientation (geometry) information from one image to
another.
If you run this and copy the info from the original image to the
segmentation
output then that will hopefully fix the problem.
All the best,
Mark
On 31 Dec 2008, at 14:19, Markus Gschwind wrote:
> Hi!
>
> You started with the same image by using two different tools, right?
> Appearently there is a flip somewhere. I'd try to find out where and
> why ( look at the intermediate stades with fslhd)
>
> Concerning the change of orientation: This is a difficult matter and
> I would try to avoid to do it manually. See many posts about
> fslswapdim and fslreorient on the list.
>
> Cheers,
> Markus
>
> 2008/12/30 siamak <[log in to unmask]>
> Hi,
>
> When I overlay the images, they don't have any overlap. It seems
> that a
> proper rerotation is required to align them. One of them is LPI the
> other is RPI.
> However, I changed both of them to RPI, but the problem exists.
>
> Cheers,
> Siamak
>
>
>
> --
> Dr. med. Markus Gschwind, M.D.
> Laboratory for Neurology and Imaging of Cognition
> Dept of Neurosciences
> University Medical Center (CMU)
> 1 Michel-Servet - 1211 GENEVA - CH
>
> Tel 0041 (0) 22 379 5324
> Fax 0041 (0) 22 379 5402
> email: [log in to unmask]
> http://labnic.unige.ch
|