Dear Dani,
On 24 Aug 2018, at 19:18, Dani Berge wrote:
> Hi everyone
>
> First of all, I would like to thank Christian Gaser for this amazing
> toolbox and for responding the email list,
> I have tried both strategies that Christian Gaser suggested. Both of
> them
> work perfectly:
>
> - The first option, if you run CAT12 in expert mode (type
> cat12('expert')
> in matlab, go to segmentation and there will be more options
> available. One
> of them is to save atlas maps in subject space. The only drawback is
> that
> you have to run segmentation again (I could not find another way).
> Then you
> will have, at each subject folder, a new folder called mri_atlas where
> the
> specified atlas in subject space (neuromorphometrics, ibpa40, .. the
> ones
> that you also specify in the segmentation options.) have been saved.
> Then
> you are ready to apply whatever roi mask you want to your subject
> (using
> imcalc or fslmaths you can extract each roi and apply it).
> - The second option (map your FDG or fMRI data to surface), also works
> fine
> following Chrisitan Gaser instructions, but, as far as I know, you
> cannot
> get the surface atlas in subject space. In my case, I was interested
> in
> combining a group of ROI (for example, total grey matter), and I
> managed to
> do this creating a new annot file (they are in cat12/atlas_surfaces/).
> Basically, using the function cat_io_FreeSurfer in matlab, I used the
> aparc_a2009s.freesurfer.annot file as a model, and modified the
> "label" mat
> variable to match the vertex with the labels you want to join with the
> new
> label. You have to look at the "colortable" mat variable to find out
> to
> wich roi corresponds to each label. Then run the "Extract ROI Data"
> functions and you will have for each subject, a 'rois' folder with a
> csv
> file with the average values for each ROI you defined.
There will be a more simpler option available in the next releases. This
option will allow to extract ROI-values from (r)fMRI or PET-data using
the surface atlases in native space in one step. Here, the Human
Connectome Project (HCP) Multi-Modal Parcellation atlas and the
Local-Global Intrinsic Functional Connectivity Parcellation by Schaefer
et al. will be particularly helpful.
Best,
Christian
>
> Hope this may help some one.
>
> Dani Bergé
>
> Missatge de Christian Gaser <[log in to unmask]> del dia dv.,
> 24
> d’ag. 2018 a les 2:23:
>
>> Dear Alain,
>>
>> you can try to save the atlas maps in native (subject space) in
>> expert
>> mode of CAT12:
>> cat12('expert')
>>
>> Then you can use these maps to extract your ROI values for
>> volume-based
>> atlases.
>>
>> However, there is again an option to also use surface-based atlas
>> ROIs:
>> - Co-register T1 to your FDG data
>> - Run CAT12 with the T1 and also enable surface processing
>> - Map unsmoothed FDG data in native space to surfaces using "Map
>> Volume
>> (Native Space) to Individual Surface"
>> - Extract ROI-based surface values
>>
>> There will be in future a less complicated approach to map fMRI or
>> PET
>> data to the native surfaces and extract ROI values in one approach.
>> This
>> might be especially helpful for rfMRI data and can better deal with
>> multiple files. Please keep in mind that for the surface ROIs no
>> information about the cerebellum and sub-cortical areas can be
>> obtained.
>>
>> Best,
>>
>> Christian
>>
>> On Thu, 23 Aug 2018 15:39:45 +0100, Alain Imaging <
>> [log in to unmask]> wrote:
>>
>>> Hi everyone,
>>>
>>> I would like to use CAT12 in order to perform ROIs analysis of PET
>>> data.
>>> Recently, I read this thread (
>> https://www.jiscmail.ac.uk/cgi-bin/webadmin?A2=ind1807&L=spm&P=R51199&1=spm&9=A&J=on&d=No+Match%3BMatch%3BMatches&z=4)
>> in which Christian Gaser suggested two different method to use cat12
>> for
>> PET data. However, the original poster of that question wanted to
>> perform
>> voxel-wise analysis on its PET data. On the other hand, i would like
>> to
>> extract FDG values (in subject space) from the set of ROIs present in
>> the
>> atlases that CAT uses.
>>> The idea is to linearly register T1 and PET images, than, if this is
>> possible, using CAT to extract the values not only GM and WM values,
>> but
>> also FDG ones. If there is no automatic way of doing this, I still
>> reasoned
>> that at a certain point, CAT must register its atlases to the T1
>> images,
>> since to my understanding the ROIs values extraction is done in the
>> subject
>> space. Is it possible to access this registration and use them for
>> FDG
>> values extraction ?
>>> Thanks in advance for any suggestion
>>> Best
>>>
>>> Alain
>>
