Hi,
well, as with any option, using it this or that way will have an impact
on the results (it should rather give you pause if it doesn't :)
The conventional way of modulation (affine + non-linear) will preserve
the original sum of voxel intensities; i.e., if you sum up the voxel
values in the native space and the "fully" modulated normalized image,
they will be identical. This is usually more easy to interpret in my
opinion. The idea from Christian was to account for global differences
only, for specific reasons I now don't want to go into. If you can
follow these arguments and find them valid, do so. If you don't, don't.
What is "better" in your dataset ultimately is your decision, but as you
said, you will have to interpret the results differently.
Cheers,
Marko
Vladimir Bogdanov wrote:
> Dear Marko and Ellen,
>
> Thank you very much for your advice!
>
> After I tried other writhing option ("linear & non-linear") the results show that what you suggested appears to be true.
>
> But if two options of processing give quite different results, which one is the optimal approach to generate GM modulated images to check for local differences in GM volume between the groups?
>
> Is it classic VBM modulation linear & non-linear?
>
> Or, maybe, both writing modulated normalized "non-linear only" and "linear & non-linear" can be used, but the interpretation of the possible between-group differences is somewhat different?
>
> Thank you a lot for sharing some thoughts on this!
>
> Sincerely yours,
> Vladimir
>
>
>> ________________________________
>> From: Marko Wilke <[log in to unmask]>
>> To: [log in to unmask]
>> Sent: Tuesday, December 1, 2015 3:11 PM
>> Subject: Re: [SPM] Question on DARTEL vs VBM: different outcome
>>
>>
>> Hello Vladimir,
>>
>> you have actually provided the clue to your problem by faithfully
>> listing your options, which is highly commendable:
>>
>>
>>
>>
>>
>>> I did DARTEL estimation of GM volume as it is described in the SPM8
>>> manual (Chapter 41, Using DARTEL).
>>> Other approach was using VBM toolbox, as described in “VBM8--‐Toolbox
>>> Manual” using default settings for writing options (Modulated
>>> normalized–non linear only).
>>
>> The "modulated non-linear only" option in the VBM8 toolbox will not do a
>> "standard" correction for the overall scaling occurring during linear
>> and non-linear spatial normalization, but will only correct for the
>> latter. Hence, the scaling resulting from the linear / affine
>> transformation will not be accounted / corrected for. The average
>> scaling introduced by the affine procedure in healthy adult brains is
>> about 1.4, which is likely about the factor that your results differ by.
>>
>> If you choose the classic modulation in VBM8 (linear & non-linear), then
>> the results should be closer (if not identical, see some discussions
>>from the last weeks).
>>
>> Cheers,
>> Marko
>> --
>> ____________________________________________________
>> PD Dr. med. Marko Wilke
>> Facharzt für Kinder- und Jugendmedizin
>> Leiter, Experimentelle Pädiatrische Neurobildgebung
>> Universitäts-Kinderklinik
>> Abt. III (Neuropädiatrie)
>>
>> Marko Wilke, MD, PhD
>> Pediatrician
>> Head, Experimental Pediatric Neuroimaging
>> University Children's Hospital
>> Dept. III (Pediatric Neurology)
>>
>> Hoppe-Seyler-Str. 1
>> D - 72076 Tübingen, Germany
>> Tel. +49 7071 29-83416
>> Fax +49 7071 29-5473
>> [log in to unmask]
>>
>> http://www.medizin.uni-tuebingen.de/kinder/epn/
>>
>> ____________________________________________________
>>
>>
>
>
--
____________________________________________________
PD Dr. med. Marko Wilke
Facharzt für Kinder- und Jugendmedizin
Leiter, Experimentelle Pädiatrische Neurobildgebung
Universitäts-Kinderklinik
Abt. III (Neuropädiatrie)
Marko Wilke, MD, PhD
Pediatrician
Head, Experimental Pediatric Neuroimaging
University Children's Hospital
Dept. III (Pediatric Neurology)
Hoppe-Seyler-Str. 1
D - 72076 Tübingen, Germany
Tel. +49 7071 29-83416
Fax +49 7071 29-5473
[log in to unmask]
http://www.medizin.uni-tuebingen.de/kinder/epn/
____________________________________________________
|