.png){kind=logo}
Microsoft researchers have developed an artificial intelligence tool that diagnoses patients step by step, much like doctors investigate medical conditions. According to a study published in the journal Science on Thursday, the system was more than twice as accurate as general physicians when solving complex medical cases.
The tool uses an AI system called an ‘orchestrator’ to coordinate diagnostic reasoning, decide which questions to ask, and determine which examinations and tests may be needed.
Unlike existing medical AI models that analyze a patient’s symptoms, laboratory results, and imaging studies together to generate a diagnosis, Microsoft’s system follows a sequential process.
The orchestrator first identifies relevant questions about the patient’s medical history and determines which physical-exam findings matter. It then decides which tests to order, evaluates the results, and requests further investigations when necessary.
Throughout the process, the system updates its working hypotheses and plans as new information becomes available, the researchers said.
The approach reflects how doctors gather information, assess possible explanations, and refine their conclusions before arriving at a diagnosis.
Also Read: Startup News Today: HealthTech Firm Gabify Secures $175,000 for AI Diagnosis Tools
Researchers tested the system on more than 300 unusual and challenging medical cases. The AI tool was more than twice as accurate at investigating patients and determining diagnoses as general physicians who used textbooks and the internet.
The system also outperformed human physicians on scenarios drawn from a major database of emergency department admissions.
Researchers said traditional assessments of medical AI often rely on multiple-choice questions. Such tests may not adequately reflect real-world clinical practice, where doctors must decide what information to gather and which investigations to pursue.
“AI has become very good at getting a 100% pass rate on those,” noted Dr Dominic King, corporate vice president of the health unit at Microsoft AI.
“But of course, that’s not real-world. No patient ever came to see me with a nice summary,” he added.
The study also found that diagnostic work-ups designed by the orchestrator cost less than those conducted by human doctors.
Harsha Nori, head of AI at Microsoft AI Health, said the evaluation considered not only diagnostic accuracy but also how carefully and efficiently the system approached the process.
“We’re not just measuring the accuracy of the systems. We’re also measuring how judiciously and carefully it goes through a diagnostic work-up,” Nori said.
The findings suggest AI could help clinicians investigate complicated cases and make more targeted decisions about additional tests.
Microsoft researchers are preparing to test the tool with Beth Israel Deaconess Medical Center in Boston. The next stage will examine the system’s ability to triage patients in emergency department waiting rooms. The research will explore how the technology could support decisions about which patients need attention first.
The latest findings mark a further step in Microsoft’s work on AI-assisted medical diagnosis. However, the study’s results concern diagnostic performance in evaluated cases and do not, by themselves, establish how the system would perform across routine clinical practice.
