Hi Peter,

lurking, indeed ;)

In any case, you could use the CerebroMatic to generate TPMs for your
population. The upper age range is, I think, 900 months, so that may be
a problem, but in general, that might be the most straightforward way.
As you have noticed, the correction is built-in.

In addition to the CerebroMatic generating custom TPMs, I have now used
it to generate custom DARTEL templates in the same fashion. I.e., you
specify the demographic parameters of your population and let the
toolbox do the rest, resulting in your very own custom DARTEL templates.

There is a paper out on this just recently, appropriately called "A
spline-based regression parameter set for creating customized DARTEL MRI
brain templates from infancy to old age" which is freely available at
https://doi.org/10.1016/j.dib.2017.12.001

Now that the advertising is over, making tissue volumes sum to 1 is
actually not that hard if you still want to do it yourself. You need to
ensure that the sum of each voxel across all classes is 1, which you can
achieve (after removing 0's) by summing over all classes as in

V = spm_vol(spm_select);
vol = spm_read_vols(V);
vol(vol==0) = eps;
corr = ones([size(vol,1) size(vol,2) size(vol,3)]) - sum(vol, 4);

and then you need to incorporate that information into the existing
volume, by combining it with the last class (which is a bit dirty but
since the last class is background anyway, it shouldn't matter), as in

vol(:,:,:,end) = vol(:,:,:,end) + corr;

upon which you now only need to write the file, as in

[p, nm, e, ~] = spm_fileparts(V.fname);
V.fname = [p filesep nm '_corrected' e];
spm_write_vol(V, vol);

or something along those lines. You may want to check that this actually
works because it's already late for me, but in principle, it should.

Hope this helps,
Marko

Stewart, Peter schrieb:
> Hi Everybody,
>
>
>
> Thank you all in advance for the invaluable information on this list. I
> have learned so much from reading all your conversations. I would be
> greatful for any help that you can give me with a project I'm working
> on. Marko, I'm hoping you are lurking out there because I am sure you
> know the answer. The situation is this....
>
>
>
> I had the idea to create a set of custom TPMs and DARTEL templates for
> older individuals with mild cognitive impairment and mild Alzheimer's
> disease. I had planned to use CAT12 to do priorless segmentation of a
> large number of individuals as in Wilke et al.'s excellent Cerebromatic
> paper. Then, I had planned to average the DARTEL registered CAT12 tissue
> classes from the subjects together to create TPMs. Looking over
> preliminary results, there are many zero voxels in the resulting TPMs
> and the probabilities across the 6 tissue classes sum to 1.1 or more in
> many cases. How can I go about correcting these, such that the tissue
> probability maps I create will add to a total probability of 1 across
> the tissue classes and there are no zeros? I know that this must be the
> case to use TPMs with SPM. I have been going through the cerebromatic
> code to see how this was achieved by Wilke but I'm having a hard time
> understanding it. Thanks in advance, I appreciate the help.
>
>
>
> Regards,
>
> Peter
>
>
>
> Peter V. Stewart, Psy.D. | Staff Neuropsychologist | Assistant Director:
> Clinical Neuropsychology Postdoctoral Training Program | Neuropsychology
> Internship Coordinator | Geisinger Health System |
> [log in to unmask] <mailto:[log in to unmask]> et al.,
>
>
> IMPORTANT WARNING
> ------------------------------------------------------------------------
>
> IMPORTANT WARNING: The information in this message (and the documents
> attached to it, if any) is confidential and may be legally privileged.
> It is intended solely for the addressee. Access to this message by
> anyone else is unauthorized. If you are not the intended recipient, any
> disclosure, copying, distribution or any action taken, or omitted to be
> taken, in reliance on it is prohibited and may be unlawful. If you have
> received this message in error, please delete all electronic copies of
> this message (and the documents attached to it, if any), destroy any
> hard copies you may have created and notify me immediately by replying
> to this email. Thank you. Geisinger Health System utilizes an encryption
> process to safeguard Protected Health Information and other confidential
> data contained in external e-mail messages. If email is encrypted, the
> recipient will receive an e-mail instructing them to sign on to the
> Geisinger Health System Secure E-mail Message Center to retrieve the
> encrypted e-mail.
>

--
____________________________________________________
Prof. Dr. med. Marko Wilke
  Facharzt für Kinder- und Jugendmedizin
  Leiter, Experimentelle Pädiatrische Neurobildgebung
  Oberarzt der Abteilung Neuropädiatrie
  Universitäts-Kinderklinik

Marko Wilke, MD, PhD
  Pediatrician
  Head, Experimental Pediatric Neuroimaging
  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/
____________________________________________________