Hacker News new | ask | show | jobs
by reubens 1189 days ago
There is none. But if the multimodal model is exposed to enough medical knowledge, it may be able to interpret images without specific training
1 comments

Labelling data is easier, I think. It will just take a while...