A benchmark of 1,063 multiple-choice visual questions drawn from the
NEJM Image Challenge (October 2005 – April 2026),
a weekly clinical image quiz published by the New England Journal of Medicine.
Each case presents a medical image and asks the reader to identify the diagnosis,
finding, or underlying process from five answer choices. The benchmark covers a
broad range of specialties including radiology, dermatology, ophthalmology… See the full description on the dataset page:
https://huggingface.co/datasets/medicalsphere/nejm-bench.