I'm focussing on learning / classifieung from 3D (specially CAD) Data. One interessting aspect, is to apply xAI Methods upon these networks, in order to understand, why these networks fond a certain feature. Therefore I am also quite curious, if this works with MRI scans too.
There was a challenge in Kaggle to find Intracranial Hemorrhage from RSNA(Radiological Society of North America) and they have a really big data of high quality MRIs. Also, you can see and learn how the competitors have classified the data and coded the model.
Here is the link: https://www.kaggle.com/c/rsna-intracranial-hemorrhage-detection/notebooks