Hi Sophie,
> Greetings helpful SPM community,
We will see about the helpful ;)
> After many helpful emails, we were able to analyze and decipher our
> VBM results and submit a manuscript.
Congratulations :)
> One of the reviewer comments were that:
>
> "A further point is which kind of template has been used for
> normalisation. If they have used the standard template provided by
> MNI they have to be aware of the fact that this template is
> completely symmetric.
Interesting comment. If I, however, call up the grey.nii from the
spm5/tpm, which is the gray matter map used for internal normalization
during unified segmentation, it does not seem to be symmetric. In fact,
if I go to voxel location 52.0 16.0 46.0, there clearly is the asymmetry
in the occipital lobe that has been described in several papers, not
least among them the famous Good et al, 2001. The same seems to be true
for the T1.nii in the templates-directory (which only comes into play if
specifically used during spatial normalization, classical approach). If
I finally check spm_templates.man, I get
% /TPM/ DIRECTORY
% Contains Tissue probability maps provided by the International
% Consortium for Brain Mapping, John C. Mazziotta and Arthur W. Toga.
% http://www.loni.ucla.edu/ICBM/ICBM_TissueProb.html
%
% The 452 subjects were between 20 and 32 years old, and were
% approximately 50-50 male/female.
%
% The original datasets were affine transformed to match the MNI152
% space (by affine registration of the grey matter with the grey
% matter image in the apriori directory: smosrc=8, smoref=0,
% regtype=mni, cutoff=Inf).
No word of symmetrical templates anywhere in the document. So
bottomline, I do not know what the reviewer is talking about.
> does anyone have any comments on this?
To me, there seems to be a misunderstanding as the data coming with spm
is clearly not symmetrical, period.
> Should I make my own templates and then run the data again? if so, how
> do I make these templates? segment T1's, then average all subjects
> (both groups) using imcalc. do I use native space (c1, c2 etc) or
> normalized (wc1) or modnor (mwc1) files for template?
My answers would be no, and don't ask (there have been a great number of
threads on how to best make custom templates, and most of them, luckily,
date back to spm99/2 days).
> thanks infinitely,
De nada.
Marko
--
=====================================================================
Marko Wilke (Dr.med./M.D.)
[log in to unmask]
Universitäts-Kinderklinik University Children's Hospital
Abt. III (Neuropädiatrie) Dept. III (Pediatric neurology)
Hoppe-Seyler-Str. 1, D - 72076 Tübingen
Tel.: (+49) 07071 29-83416 Fax: (+49) 07071 29-5473
=====================================================================
|