A Specialized Large Language Model for Clinical Reasoning and Diagnosis in Rare Diseases
By: Tao Yang , Dandan Huang , Yunting Lin and more
Potential Business Impact:
Helps doctors find rare diseases faster.
Rare diseases affect hundreds of millions worldwide, yet diagnosis often spans years. Convectional pipelines decouple noisy evidence extraction from downstream inferential diagnosis, and general/medical large language models (LLMs) face scarce real world electronic health records (EHRs), stale domain knowledge, and hallucinations. We assemble a large, domain specialized clinical corpus and a clinician validated reasoning set, and develop RareSeek R1 via staged instruction tuning, chain of thought learning, and graph grounded retrieval. Across multicenter EHR narratives and public benchmarks, RareSeek R1 attains state of the art accuracy, robust generalization, and stability under noisy or overlapping phenotypes. Augmented retrieval yields the largest gains when narratives pair with prioritized variants by resolving ambiguity and aligning candidates to mechanisms. Human studies show performance on par with experienced physicians and consistent gains in assistive use. Notably, transparent reasoning highlights decisive non phenotypic evidence (median 23.1%, such as imaging, interventions, functional tests) underpinning many correct diagnoses. This work advances a narrative first, knowledge integrated reasoning paradigm that shortens the diagnostic odyssey and enables auditable, clinically translatable decision support.
Similar Papers
Evaluating Large Language Models on Rare Disease Diagnosis: A Case Study using House M.D
Computation and Language
Helps computers guess rare diseases from stories.
Quantifying the Reasoning Abilities of LLMs on Real-world Clinical Cases
Computation and Language
Tests AI doctors' thinking for better patient care.
EHR-R1: A Reasoning-Enhanced Foundational Language Model for Electronic Health Record Analysis
Computation and Language
Helps doctors understand patient health records better.