Actively Recruiting
Multicenter Randomized Trial of AI-Assisted Diagnostic Support for Suspected Rare or Unresolved Diseases
Led by Peking Union Medical College Hospital · Updated on 2026-10-13
1056
Participants Needed
12
Research Sites
21 weeks
Total Duration
AI-Summary
What This Trial Is About
Researchers are investigating whether an AI-powered large language model can help doctors improve the diagnosis of rare or difficult-to-diagnose diseases. This trial compares the use of AI-assisted diagnostic support with standard clinical care to see if it improves diagnostic accuracy, speeds up diagnosis, and affects health-economic outcomes for patients with suspected rare diseases. The study is a multicenter randomized controlled trial involving physicians and patients across various hospitals. Participants are randomly assigned to one of two groups one where doctors use the AI system during patient consultations to organize clinical information, analyze cases, and suggest possible diagnoses, and another group where doctors follow standard diagnostic procedures without AI assistance. Use of other generative AI tools is not allowed in either group. The trial also includes an economic evaluation to track healthcare resources and costs related to the diagnostic process. Participants will have a first consultation with their physician, during which various measures are recorded, including the duration of the consultation and experiences reported by both patients and doctors. Outcomes such as correct diagnostic yield, diagnostic accuracy, time to diagnosis, and appropriate genetic testing recommendations will be evaluated over approximately eight weeks following the initial visit. The study uses blinded independent experts to assess the diagnostic outcomes, and participants are expected to complete structured follow-up during this period.
CONDITIONS
Brief Title
AI-assisted Rare Disease Diagnosis
Research Team
S
Shuyang Zhang, MD, PhD
Not the Right Trial for You?
Explore thousands of other clinical trials that might be a better match.
Sign up to get personalized trial recommendations delivered to your inbox.
Already have an account? Log in here