Hello,
If I understand correctly, you want to create a larger reference space but maintain the co-ordinate system, for this you can use fslroi to expand the image e.g. for the 2mm standard image MNI152_T1_2mm.nii.gz:
fslroi MNI152_T1_2mm.nii.gz output -10 111 -10 129 -10 111
will create an output image expanded by 10 voxels in each direction, but with all anatomical locations maintaining their mm co-ordinates.
Hope this helps,
Kind Regards
Matthew
--------------------------------
Dr Matthew Webster
FMRIB Centre
John Radcliffe Hospital
University of Oxford
> On 29 Nov 2017, at 17:38, Alexis Guyot <[log in to unmask]> wrote:
>
> Dear FSL team,
>
> I have been trying to register an isotropic (0.8*0.8*0.8 mm3) whole-brain T2-weighted source volume with an anisotropic (0.3*1.2*0.3 mm3) T2-weighted volume covering the brain partially and centred on a region comprising both hippocampi.
> To get my registered volume to cover the whole extent of the brain, I produced a nifti reference volume encompassing the target volume as follows:
> 1) the reference volume has a higher number of voxels than the target volume in each direction
> 2) the reference volume has the same affine matrix as the target volume, except that it has nonzero values in the fourth (translation) row
> Somehow, after I used '-applyxfm -ref [reference volume]' FSL flirt seemed to ignore the translation row and returned a registered volume which was not aligned with the target image.
>
> Is this the expected behaviour? If so, could you please recommend an alternative way to produce a reference that encompasses the target volume?
>
> Thanks for your help,
> Alexis
|