One of the developers here. The paper links to an earlier model from a different group that could only interpret X-rays of specific body parts. Our model does not have such limitation.

However, the actual FRCR 2B Rapids exam question bank is not publicly available and the FRCR is unlikely to agree to release them as this would compromise the integrity of their examination in the future- so the test used are mock examinations, none of which have been provided to the model during training.

Interesting, is your model still based on radiographs alone, or can it look at cross-sectional imaging as well?

This current model is radiographs alone. The FRCR 2B Rapids exam is based on only radiographs.

Agreed, updated at https://harrison.ai/harrison-rad-1/

Great. Kudos to you for being open to the feedback!

No problem! Thanks for picking it up.