Microsoft says a diagnostic AI beat generalist doctors on hard simulated cases
A Science paper puts MAI-DxO at 83 percent correct, against 36 percent for individual primary-care doctors. Specialists were not in the comparison.
A Microsoft diagnostic system correctly diagnosed complex medical cases in a simulation more than twice as often as generalist physicians, according to a study published Thursday in Science, TechCrunch reported. Mustafa Suleyman, CEO of Microsoft AI, called the result a step toward what he described as medical superintelligence.
The system, MAI-DxO, coordinates several AI agents that question a patient, order tests and debate a diagnosis. On complex cases from the New England Journal of Medicine it was correct 83 percent of the time. Primary care doctors and general internists, working alone with reference materials, were correct 36 percent of the time. Cases this difficult are usually handled by specialists or teams, who were not part of the comparison.
From ARQS Sponsored
For health and life-science teams, ARQS offers Design Workbench, TradeHound, Health Intelligence and Website audit. In the ARQS Store, Menu Maker — 1 Month Access, Lead Scout Local — 1 Month Access, Inbox Triage Bot — 1 Month Access are among the newest tools available.
- AI Academy courses
- ARQS Store
- Northline Technology Institute
- ROI Intelligence
- Business network
- Advertise with ARQS
- AI chatbot hosting
- Website packages
- Reputation engine
- Resume service
- ARQS Sudoku Logic Lab
- AICL 104 — Regulatory Informatics & Unstructured Data
- AICL 102 — Mathematical Foundations of Machine Learning