Hello,
with regard to the first part, is using a custom template in children
aged 8-12 a good idea, my reply would be "definitely".
With regard to the second question ("how many subjects do I need for my
own templates"), I refer to Figures 6 & 7 in the paper at
https://doi.org/10.3389/fncom.2017.00005 The simulations therein (to me)
conclusively demonstrate that there is a substantial difference between
priors generated from 25, 50, or 100 subjects, and a gold standard, and
that this difference is substantially lower for com-priors.
Cheers
Marko
Ariadna Albajara Sáenz schrieb:
> Dear Marko,
>
> Thank you for your answer. My sample has 73 children (56 after excluding
> the children with motion). Do you think in this case it is appropriate
> to use a customized DARTEL template ?and customized TPMs?
>
> You are right, I will check out Cerebromatic! ;)
>
> Thanks again,
>
> *Ariadna Albajara Sáenz*
> [log in to unmask] <mailto:[log in to unmask]>_
> UR2NF - Neuropsychology and Functional Imaging Research Group
> CRCN - Center for Research in Cognition and Neurosciences
> http://crcn.ulb.ac.be/
> *
> *
> UNIVERSITÉ LIBRE DE BRUXELLES, Avenue F. Roosevelt 50, CP 151, 1050
> Brussels (Belgium).
> Office: DB10-237
>
>
> El vie., 26 jul. 2019 a las 15:47, Marko Wilke
> (<[log in to unmask]
> <mailto:[log in to unmask]>>) escribió:
>
> Hi,
>
> due to Friday afternoon and the current heatwave, I will only answer in
> very general terms:
>
> - I cannot judge whether using your own DARTEL template is a good idea
> as I do not know how large your group is (my hunch is it should be
> really large for this to be a good idea)
>
> - I am not a believer that a given space (in this case, MNI) is
> automatically better than another; if you use custom templates, I find
> it equally legitimate to describe your results based on where they are
> in your template
>
> - finally (pardon my bluntness), lack of familiarity with a given
> approach is not a strong scientific argument (it also hurts my pride as
> I invested several hours into writing a manual for COM)
>
> Hope this helps
> Marko
>
> Ariadna Albajara Sáenz schrieb:
> >
> > Dear Marko and Helmut,
> >
> > Thank you for the feedback. I wanted to check if the steps I
> followed
> > are correct in case these could explain my results.
> > After checking data quality and excluding participants with
> > artefacts/motion, I followed the following steps (taking into
> account
> > that I have a pediatric sample aged 8-12 and that I am using TOM8
> and
> > DARTEL since I am more familiar with these than Cerebromatic):
> >
> > 1. I created customized TPMs using TOM8.
> >
> > 2. Then I created a DARTEL template using segmentation (grey matter
> > ->Dartel export affine, white matter -> Dartel export affine) and
> here I
> > entered already the previously created TOM8 TPMs.
> >
> > 3. Then I used "DARTEL tools--> normalise to MNI space" (I did this
> > because it is in the Manuel although this creates a ".mat" file
> called
> > "Template_6_2mni.mat" and the images "smwrp1_XXXX-affine.nii" and
> > "smwrp2_XXXX-affine.nii", which I do not use in the following
> steps SO I
> > DON'T KNOW IF THIS STEP IS NECESSARY?).
> >
> > 4. Then I did "DARTEL ICBM --> population to ICBM registration"
> where I
> > entered the file "Template_6" created in the step 2.
> >
> > 5. After this, I did "SPM-> Util--> deformations" where I entered
> the
> > "y_Template-6-2mni.nii" file created in the step 4 and under
> "Apply to"
> > I entered the Templates 0 to 6 created in step 2. This creates
> the files
> > wTemplates 0 to 6
> >
> > Finally I do segmentation entering the ORIGINAL T1 images but
> this time
> > entering the TPMs created with TOM8 in "Tissue Probability Map"
> and the
> > file "wTemplate_1.nii" in Dartel template. Is this correct? is it
> > correct to enter the original T1 images here?
> >
> >
> > Does this seem correct?
> >
> > I was also wondering if the subsequent statical maps would be in MNI
> > space or not.
> >
> > Thanks a lot for your help.
> >
> > *Ariadna Albajara Sáenz*
> > [log in to unmask] <mailto:[log in to unmask]>
> <mailto:[log in to unmask] <mailto:[log in to unmask]>>_
> > UR2NF - Neuropsychology and Functional Imaging Research Group
> > CRCN - Center for Research in Cognition and Neurosciences
> > http://crcn.ulb.ac.be/
> > *
> > *
> > UNIVERSITÉ LIBRE DE BRUXELLES, Avenue F. Roosevelt 50, CP 151, 1050
> > Brussels (Belgium).
> > Office: DB10-237
> >
> >
> > El jue., 25 jul. 2019 a las 12:54, MRI More (<[log in to unmask]
> <mailto:[log in to unmask]>
> > <mailto:[log in to unmask] <mailto:[log in to unmask]>>>) escribió:
> >
> > Dear Ariadna,
> >
> > The glassbrain image looks a bit noisy, which doesn't have to
> mean
> > anything though. Based on the location, the MRI image seems to
> > indicate GM differences in the caudate nucleus and the basal
> ganglia
> > more generally. Please note that parts of the basal ganglia
> and the
> > thalamus are quite difficult to identify in certain templates
> due to
> > poor GM-WM contrast.
> >
> > Did you use the Dartel or Shoot template as included in CAT12 for
> > the high-dimensional registration part? In that case, it might be
> > wise to overlay the results onto the corresponding template
> (GM-WM
> > contrast for basal ganglia should be better), or even better,
> create
> > a mean image of the normalised (bias corrected) structural
> volumes
> > of your study, or alternatively, of the normalised modulated or
> > unmodulated GM files as obtained from CAT12 preprocessing.
> IMO it's
> > always preferable to map results on some image derived from
> your own
> > data. If there were some issues during the
> > segmentation-normalisation, resulting in enlarged ventricles
> > relative to the templates / priors, it would obviously be
> misleading
> > to rely on a mapping onto those templates.
> >
> > Best regards
> >
> > Helmut
> >
>
> --
> ____________________________________________________
> Prof. Dr. med. Marko Wilke
> Facharzt für Kinder- & Jugendmedizin, Neuropädiater
> Leiter, Experimentelle Pädiatrische Neurobildgebung
> Geschäftsführender Oberarzt der Abt. Neuropädiatrie
> Universitäts-Kinderklinik
>
> Marko Wilke, MD, PhD
> Pediatrician and Pediatric Neurologist
> Head, Experimental Pediatric Neuroimaging
> Senior Consultant in Pediatric Neurology
> University Children's Hospital
>
> Hoppe-Seyler-Str. 1
> D - 72076 Tübingen, Germany
> Tel. +49 7071 29-83416
> Fax +49 7071 29-5473
> [log in to unmask]
> <mailto:[log in to unmask]>
>
> http://www.medizin.uni-tuebingen.de/kinder/epn/
> ____________________________________________________
>
--
____________________________________________________
Prof. Dr. med. Marko Wilke
Facharzt für Kinder- & Jugendmedizin, Neuropädiater
Leiter, Experimentelle Pädiatrische Neurobildgebung
Geschäftsführender Oberarzt der Abt. Neuropädiatrie
Universitäts-Kinderklinik
Marko Wilke, MD, PhD
Pediatrician and Pediatric Neurologist
Head, Experimental Pediatric Neuroimaging
Senior Consultant in Pediatric Neurology
University Children's Hospital
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/
____________________________________________________
|