Reproduce “Ontology-guided multi-agent reasoning for explainable differential diagnosis” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Roxana Ramona Szomiu, George Ursu, Adrian Groza
MedMAS-DDx turns incomplete symptom notes into ranked differential diagnoses by mapping symptoms to medical relationships, asking questions about missing findings, and updating rankings with deterministic code. Across 198 synthetic Q4Dx cases, follow-up raised Top-1 accuracy from 18.7% to 21.2% and Top-3 from 25.8% to 27.3%. Yet Top-3 fell at 50% completeness, and simulated-patient results do not establish real-world performance.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.