For many Dicom converters (such as the one in SPM), the 0,0,0
coordinate will be placed at the centre of the scanner's magnetic
field. It is rare for the AC to be positioned at exactly the centre
of the scanner, so it's hardly surprising that some translation is
needed.
You should not need to write out any reoriented images. Just make
sure the matrix in the header is good, which is what SPM changes when
you reorient an image via the Display button.
Best regards,
-John
On 22 March 2011 14:18, Ben Becker <[log in to unmask]> wrote:
> Dear SPMers,
>
> I have a strange dicom to nifti conversion issue. When I load our epi and t1 images (converted from dicom to nifti) and the spm epi-template and the single_subject T1 our images are shifted forward. That means if I use the reorient function I have to shift our images 55mm backwards (forward = -55mm) to match the templates.
>
> I already tried two different dicom2nifti conversion applications (mriconvert & mricron) – images remained shifted.
>
> My questions:
>
> 1. For the present study
> If I reorient the raw-nifti-images using the reorient function (like described above) & write out the re-oriented images: would it be reasonable to use the reoriented images for the further analyses?
>
> 2. For future studies
> What might have caused the forward shift in our images?
>
> Thanks & kind regards,
>
> Ben
>
|