RareLens: AI System for Rare Disease Diagnosis
RareLens is an AI system designed to support clinical decision-making across the entire rare disease trajectory. It addresses the challenge that rare diseases affect 3.5% to 5.9% of the population, with over 70% of patients misdiagnosed. The system exploits variability in reasoning from heterogeneous large language models, aligning and calibrating their divergent outputs into convergent, actionable decisions at each stage of care. RareLens consists of four coordinated modules, moving beyond existing AI systems that focus on isolated stages like diagnosis and depend on downstream investigation results.
Key facts
- Rare diseases affect 3.5% to 5.9% of the population.
- More than 70% of rare disease patients are misdiagnosed.
- Many patients endure years of evaluation before diagnosis.
- Existing AI systems address isolated stages of care, overwhelmingly diagnosis.
- RareLens supports clinical decision-making across the entire rare disease trajectory.
- RareLens aligns and calibrates divergent reasoning from heterogeneous large language models.
- The system produces convergent, actionable decisions at each stage.
- RareLens consists of four coordinated modules.
Entities
—