Thank you very much, John! It indeed worked after reslicing the lesion
mask to the T1 image.


-----Original Message-----
From: John Ashburner [mailto:[log in to unmask]] 
Sent: Thursday, March 26, 2009 9:05 AM
To: [log in to unmask]
Subject: Re: [SPM] Normalization error

I'm guessing that MRIcron was used to define the lesion mask.  In most
cases, the resulting masks will be written out such that the order of
the voxels is reversed in the x direction, with a corresponding flip
encoded in the matrices in the NIfTI headers.  Most parts of SPM will
see the data in alignment with each other, but for the lesion masking,
the voxels themselves have to correspond.

I suggest maybe using the reslice only option of Coreg to reslice the
lesion mask so that the voxels are aligned with the T1w.

Best regards,

On Wednesday 25 March 2009 20:36, Xu, Ben (NIH/NINDS) [E] wrote:
> Hi,
> I got the following error message when normalizing a T1 image with a 
> lesion mask (skull scripted with 0s for the lesion part and 1s for the

> normal tissue). The orientation of the mask and the T1 images seems to

> be the same. What could be the problem?
> Thanks for your help!
> Ben
>  --------------------------
> Running "Normalise: Estimate & Write"
> Smoothing by 0 & 8mm..
> Coarse Affine Registration..
> Fine Affine Registration..
> Error running job: Error using ==> spm_affreg at 67 Source and its 
> weighting image must have same orientation In file 
> "/usr/local/spm5/spm_affreg.m" (v1160), function "spm_affreg" at line 
> 67. In file "/usr/local/spm5/spm_normalise.m" (v946), function 
> "spm_normalise" at line 182. In file
> (v1032), function "estwrite" at line 491.