search

juaml / junifer

Forschungszentrum Jülich Neuroimaging Feature Extractor
https://juaml.github.io/junifer
GNU Affero General Public License v3.0
14 stars 13 forks source link

[MARKER]: T1w/T2w ratio #176

Open LeSasse opened 1 year ago

LeSasse commented 1 year ago

Which marker do you want to include?

The T1w/T2w ratio has been proposed as a measure/estimate of cortical myelin content in the grey matter (https://pubmed.ncbi.nlm.nih.gov/23567887/) but has also been used in the whole brain (https://pubmed.ncbi.nlm.nih.gov/26104102/) and the white matter (https://pubmed.ncbi.nlm.nih.gov/27178000/).

Is there any publication or available code?

Publications are linked above. It should be fairly simple: elementwise division between T1w image and T2w image. From the resultant image one can then extract spheres or parcels (or use it as a voxelwise measure.)

Do you have a sample code that implements this outside of junifer?

No response

Anything else to say?

No response

LeSasse commented 1 year ago

https://openneuro.org/datasets/ds003653/versions/1.0.0 I think this would be an interesting dataset to add as a datagrabber, to implement and test/validate this marker, as the related paper looks exactly at this T1w/T2w ratio @kaurao @synchon. Ah, but the problem is, I guess, they don't seem to have preprocessed derivatives, so that makes it more complicated.

fraimondo commented 1 year ago

We still do not have any dataset with T2 images. Is this a feature that we need in the short-term or something as a result of a brainstorm and the issue is here to document that?

LeSasse commented 1 year ago

this:

something as a result of a brainstorm and the issue is here to document that?