Microsoft has developed an artificial intelligence system that diagnosed complex medical conditions up to four times more accurately than generalist doctors in controlled tests, according to research published in the scientific journal Science on Thursday, October 22.
The system, called Microsoft AI Diagnostic Orchestrator (MAI-DxO), uses multiple AI agents to analyze symptoms, review medical histories, order tests, and identify diseases.
Microsoft reported that the system achieved approximately 83% accuracy in complex medical cases, compared with 36% for doctors using reference materials and 22% for those working without them.
However, the technology has only been evaluated using historical medical cases and has not yet demonstrated the same performance in real clinical settings.
How Microsoft’s AI Works
Microsoft designed the Diagnostic Orchestrator to function like a team of medical specialists working together to diagnose a patient.
Instead of immediately providing an answer, the system reviews available information, considers possible diagnoses, requests additional tests, and updates its conclusions as new evidence becomes available.
The system uses OpenAI’s GPT-5 model and coordinates several AI agents that challenge each other’s conclusions before reaching a diagnosis.
Microsoft also developed a testing framework called Sequential Diagnosis Benchmark (SDBench), which evaluates how AI systems and doctors diagnose conditions through multiple steps.
The company plans to make the benchmark available to researchers so they can evaluate other AI systems.
AI vs Human Doctors
Microsoft tested its system using 304 complex medical cases published in The New England Journal of Medicine and 325 emergency department cases from Beth Israel Deaconess Medical Center in Boston.
The results showed that the AI system performed better than generalist physicians under the tested conditions.
| Medical Cases | Microsoft’s AI | Human Doctors |
|---|---|---|
| Complex cases without reference materials | 83% | 22% |
| Complex cases with reference materials | 83% | 36% |
| Emergency department cases | 80% | 47% |
The AI system was approximately 3.7 times more accurate than doctors working without references and 2.3 times more accurate than those using reference materials.
For emergency department cases, the AI achieved 80% accuracy compared with 47% for physicians.
Microsoft also reported that versions of its system optimized for cost could reach accurate diagnoses while ordering fewer or less expensive tests.
Hospital Testing Planned
Despite the results, Microsoft acknowledged that further clinical testing is necessary before the technology can be widely deployed.
The research relied on previously documented medical cases rather than direct interactions with patients. It also compared the system with individual generalist physicians, not specialist teams.
Microsoft plans to begin testing the Diagnostic Orchestrator in real hospital settings through a partnership with Beth Israel Deaconess Medical Center in Boston.
Dr. Dominic King, Microsoft’s AI health lead, told The Verge that testing could begin within the next couple of months.
The trials will examine how the technology performs with real patients and whether it can support doctors during clinical decision-making.
Microsoft’s Medical AI Plans
The research forms part of Microsoft’s broader effort to develop what it calls “medical superintelligence,” an AI system capable of outperforming leading healthcare teams across multiple medical tasks.
The company believes AI could eventually assist with diagnosing diseases, selecting treatments, coordinating clinical care, and managing other healthcare responsibilities.
Microsoft AI CEO Mustafa Suleyman said the technology could significantly change healthcare delivery over the coming years.
However, he emphasized that human doctors should remain central to patient care, with AI helping reduce repetitive work and allowing medical professionals to spend more time with patients.
Microsoft has not announced when the Diagnostic Orchestrator will become commercially available. Any wider deployment would need to meet medical safety, regulatory, and compliance requirements.
The post Microsoft’s New AI Diagnoses Diseases Up to 4 Times More Accurately Than Doctors appeared first on ProPakistani.