Hello John,
Thank you very much for your help, this is hopefully my last question
about this procedure:
I applied new segment on the normalized T1 picture, used DARTEL
(existing template) with the templates 6 to 1 and applied backward
deformations with the resulting flowfield. After this, the images should
be in the template space, from what i understand?
As a last step, i used the first flowfield (from the first normalization
into MNI space) and made a forward deformation from the template space
to subject space.
Are these steps correct? I used check reg to compare the original T1
picture with the now several times deformed T1 picture, and it seemed
close enough. Are the resulting small differences due to interpolation
errors or is there still something wrong in the workflow?
Thank you again for your kind help!
Best regards
Robert
Am 02.02.2012 16:49, schrieb John Ashburner:
> If the data are just warped to the population average, then warping
> them back to native space should be very straightforward. The
> complication comes from the additional affine transform that is
> involved when using "normalise to MNI".
>
> For examples of using the Deformations utility, take a look at Section
> 42.4 of the SPM8 manual (Warping one individual to match another). If
> you can segment it, then you can warp your MNI-space scan to match the
> population average (via "Run Dartel (existing template)"), and then
> combine the transforms as described in the manual.
>
> Best regards,
> -John
>
> On 2 February 2012 15:35, Robert Lech<[log in to unmask]> wrote:
>> Dear SPMers,
>>
>> My problem still persists and i still wasn't able to successfully transform
>> DARTEL-normalized pictures back to subject space. I would greatly appreciate
>> any help on that matter.
>>
>> Best regards
>> Robert Lech
>>
>>
>> Am 27.01.2012 22:07, schrieb Robert Lech:
>>
>>> Thank you very much for your quick answer.
>>>
>>> I already thought that SPM should write new files, but i couldn't find
>>> any. There were none in the current Matlab directory, and there was no
>>> option in "Create Inverse Warped" to specify an output directory.
>>>
>>> Could you give me more information on warping back to original space or
>>> point me to a source with more information on that topic?
>>>
>>> Best regards
>>> Robert Lech
>>>
>>>
>>> Am 27.01.2012 16:02, schrieb John Ashburner:
>>>> The original images should remain the same. There should be new
>>>> images that get warped back to match the individual. Where they
>>>> appear will depend on what output directory you chose.
>>>>
>>>> Note that warping from MNI space to original space is a bit tricky, as
>>>> it needs more than just the Dartel flow fields. These fields map
>>>> between individual and population average spaces. An additional
>>>> affine transform is needed to map between population average and MNI
>>>> space - which will need more code to be written.
>>>>
>>>> Best regards,
>>>> -John
>>>>
>>>> On 27 January 2012 13:38, Robert Lech<[log in to unmask]> wrote:
>>>>> Dear SPM community,
>>>>>
>>>>> I recently tried to transform images from MNI space back to subject
>>>>> space,
>>>>> using the "Create Inverse Warped" tool. I put in the appropriate
>>>>> flowfield
>>>>> (which can be used for both forward and backward deformations?) and the
>>>>> images i wanted to transform.
>>>>>
>>>>> But after doing this i realized that the images were not changed at all,
>>>>> and
>>>>> they still remain in MNI space, although SPM successfully completed the
>>>>> batch without any error message.
>>>>>
>>>>> Is this the wrong tool to transform DARTEL normalized pictures back or
>>>>> am i
>>>>> missing something here?
>>>>>
>>>>> Best Regards
>>>>> Robert Lech
>>>>>
>>>>> --
>>>>> Dipl.Psych. Robert Lech
>>>>> Institute of Cognitive Neuroscience
>>>>> Department of Neuropsychology
>>>>> Ruhr-University Bochum
>>>>> GAFO 05/611
>>>>> Universitätsstraße 150
>>>>> 44801 Bochum, Germany
>>>>>
>>>>> Phone: +49(0)234-3223175
>>>>> Email: [log in to unmask]
>>>
>> --
>> Dipl.Psych. Robert Lech
>> Institute of Cognitive Neuroscience
>> Department of Neuropsychology
>> Ruhr-University Bochum
>> GAFO 05/611
>> Universitätsstraße 150
>> 44801 Bochum, Germany
>>
>> Phone: +49(0)234-3223175
>> Email: [log in to unmask]
--
Dipl.Psych. Robert Lech
Institute of Cognitive Neuroscience
Department of Neuropsychology
Ruhr-University Bochum
GAFO 05/611
Universitätsstraße 150
44801 Bochum, Germany
Phone: +49(0)234-3223175
Email: [log in to unmask]
|