The error is because SPM assumes that all data are stored in the exact same
orientation - but this is not the case: MINC images are usually stored one
way; Analyze images are stored another.
You could get around the problem by reslicing the brainmask image to match
your template, using the "reslice only" option of either Coregister, or
Realign.
Best regards,
-John
On Monday 10 October 2005 14:47, Mark Daglish wrote:
> Dear All,
>
> I am trying to use a home-made template image for spatial
> normalization. The template is an Analyze 7.5 format image. The
> bounding box is 91x109x91 - the same as the other templates. The voxel
> sizes are -2x2x2 - the same except for the flip as it's an analyze image
> not minc. The origin is at 46 64 37 - the same as the other templates.
> As far as I can see, the brainmask
> (/usr/local/spm/spm2/apriori/brainmask.mnc) is in exactly the same
> orientation as the home-made template except for the MINC/Analyze format
> difference.
>
> If I change the normalization to include default brainmask I get the
> following error:
>
> SPM2: spm_normalise_ui (v2.10) 14:39:59 - 10/10/2005
> ========================================================================
> Smoothing by 0 & 8mm..
> Coarse Affine Registration..
> Fine Affine Registration..
> ??? Error using ==> spm_affreg
> Reference images must all have the same orientation
>
> Error in ==> spm_normalise at 173
> [M,scal] = spm_affreg(VG1, VF1, aflags, M,scal);
>
> Error in ==> spm_normalise_ui>run_ui at 352
> spm_normalise(Template, subj(i).P, subj(i).matname,...
>
> Error in ==> spm_normalise_ui at 272
> run_ui(defs);
>
> ??? Error while evaluating uicontrol Callback.
>
>
> Does anyone know how I can get round this?
>
> Many thanks,
>
> Mark
|