Default anatomy & AICHA atlas

Hello Francois,

Is it possible to overlay the AICHA atlas onto the default anatomy templates housed in Brainstorm? I've tried using both ICBM152 & Colin27_BrainSuite_2016 but am running into issues with the accuracy of AICHA on cortex, even with switching the resolution of cortex (high vs low vertex #). Could this issue be due to FreeSurfer/BrainSuite generating an output that does not work well with AICHA? If so, is there a way to segment the brain in a way that works well with AICHA?

Any help would be appreciated. Thank you!

Adam

Screenshot of the ICBM cortex with AICHA overlay:

The AICHA parcellation is a volume parcellation, and should only be used as volume scouts on a volume source model:
https://neuroimage.usc.edu/brainstorm/Tutorials/TutVolSource

This mistake is common, therefore I added a new warning in the interface:
image

Hello Francois,

I appreciate your quick response and apologize for my delay.

Since my first post, our group has decided to stop trying to use AICHA for our EEG analysis and use Desikan-Killiany instead. Do you know if it is possible to export the Desikan-Killiany atlas as a nii file from brainstorm? If not, do you know of where to find it?

Thanks!
Adam

Do you know if it is possible to export the Desikan-Killiany atlas as a nii file from brainstorm?

The Desikan-Killiany atlas is generated in FreeSurfer as a surface parcellation.
A volume version is created during the recon-all segmentation pipeline, and then imported in the Brainstorm database as a volume parcellation.
To export this volume parcellation as a .nii file: right-click on the file in the subject anatomy > File > Export to file.
https://neuroimage.usc.edu/brainstorm/Tutorials/LabelFreeSurfer#Volume_parcellations

Francois,

Thank you. I was able to download a file named Deskian-Killiany_reg_volatlas.nii. This does make me have another question about this file - is this a volume or cortical parcellation? Does the "volatlas" name mean that it is a volumetric parcellation?
We would like to have a cortical parcellation for our EEG analysis.

Thank you!
Adam

This is a volume parcellation, based on the parcellation of the cortical surface in FreeSurfer.
https://neuroimage.usc.edu/brainstorm/Tutorials/LabelFreeSurfer#Cortical_parcellations

Does the "volatlas" name mean that it is a volumetric parcellation?

Yes.

We would like to have a cortical parcellation for our EEG analysis.

Are you doing your analysis in Brainstorm?
If so, why do you need to export this atlas as a .nii file?

Hi Francois,

Our subjects have brain lesions (post-stroke) and FreeSurfer does not parcellate lesioned brains well. We could manually create scouts to use within brainstorm, but this is very tedious. We know how to parcellate lesioned brains with any volume atlas within SPM, so we thought we could use the Desikan-Killiany surface-based atlas in .nii format in SPM to correctly parcellate the lesioned brain. Is there another way get correctly parcellated brains, other than manually drawing scouts?

Thanks,
Adam

If you are trying to obtain a Desikan-Killiany atlas from a lesioned brain you can't process with FreeSurfer, you could try processing it using CAT12 or FastSurfer:

Alternatively, if your initial plan was to work with the AICHA atlas, you can work with volume head models and volume scouts:
https://neuroimage.usc.edu/brainstorm/Tutorials/TutVolSource

Sorry for the response delay...