Dear Bas,
thanks for your quick answer. I got it working now.
Tank you very much,
Paul
Am 02.04.2012 10:18, schrieb Bas Neggers:
> Hi Paul.
>
> did you manually reorient the T1 scan such that it roughly matches
> your template, prior to normalization? To me it seems you started out
> with a T1 'looking' in the wrong direction, that is, looking to the
> right (where the template 'looks' to the left). Your deformed image
> still looks' to the right, that is why I have this suspicion. Such a
> big difference with the desired template cannot be handled by
> normalization, you might have ended up in a local minimum. Start out
> with an image that is already roughly in register with your template.
>
> Good luck,
>
> Bas
>
> On 04/02/2012 11:09 AM, Paul Czienskowski wrote:
>> Dear all,
>>
>> I am trying to accomplish a normalization of MR-images with a script,
>> using the SPM2-Version of spm_normalise. Unfortunately the
>> 'normalized' image does not match what I was expecting, after
>> transforming the image with spm_write_sn, but looks gravely distorted
>> (see attached image). The code I've used is approximately like the
>> following:
>>
>> mri_template =
>> spm_vol(fullfile(cd,'external','spm2','templates','T1.mnc'));
>>
>> spm_normalise(mri_template,mri_source,deformation_file);
>>
>> spm_write_sn(mri_source, deformation_file);
>>
>> I would appreciate very much, if someone has got any clue what went
>> wrong.
>>
>> Tanks in advance,
>> Paul
>>
>
>
|