Atlas and Neural Ontology

[vborghe]

@pinheirochagas is the lead on this issue 🙏
Steps:

• chose reference atlas
• establish mapping from more-vague terms to the atlas
• look for tools to cross-map between atlases
• look for tools to map from voxels to the chosen atlas

[pinheirochagas]

Hi, this tool neuroparc might be a very good solution:
https://github.com/neurodata/neuroparc.
paper:
https://www.nature.com/articles/s41597-021-00849-3.pdf?proof=t%C2%A0
I'm gonna explore it and report back.

[pinheirochagas]

To define the base atlas, we need to:

• Select the clinical papers.
• Count the frequency of atlases.
• Select the atlas that is mentioned in most studies as the base atlas.
  (probably Harvard-Oxford cortical and subcortical structural atlases or
  WFU pickatlas - labels and Broadman areas).
• Count how many other atlases are mentioned.
• If just a couple more, do a manual conversion across regions to the base atlas.
• If a variety, we must choose a method to convert from labels to MNI coordinates to map to the base atlas.
  (this will required some decisions, such as which coordinate to use, centroid?)

[vborghe]

See #12