Large Language Models for Rare Disease Diagnosis at the Undiagnosed Diseases Network | Synapse
August 22, 2025Open Access
Large Language Models for Rare Disease Diagnosis at the Undiagnosed Diseases Network
Key Points
Large language models can accurately identify diagnoses for rare diseases in patients referred to the Undiagnosed Diseases Network, improving diagnosis rates.
The study assessed clinical information of patients from the Undiagnosed Diseases Network, revealing that LLMs can enhance diagnostic accuracy.
Cohort study analyzing patient data demonstrates the ability of LLMs to assist in achieving diagnoses for complex cases.
Results support the integration of LLMs in clinical settings, suggesting benefits for diagnosing rare diseases and reducing diagnostic odyssey.
Abstract
This cohort study assesses whether large language models (LLMs) can identify the final diagnosis, based on available clinical information, for patients referred to the Undiagnosed Diseases Network.