RareDx: Controlled Knowledge Integration and Graph-Grounded Policy Optimization for Rare-Disease Diagnosis

2026-09-28

AI Summary

This study addresses the long-tail reasoning challenges in rare disease diagnosis, including incomplete phenotypes, fragmented evidence, and model bias toward common diseases. We propose KGPO, a framework built upon Qwen3.5 that unifies heterogeneous records through controlled evidence integration and maps predictions onto a disease knowledge graph. By incorporating an optimization strategy combining ontological proximity with phenotypic consistency rewards, alongside adaptive tool invocation and structured reasoning, the method effectively mitigates diagnostic hallucinations. Experiments demonstrate that KGPO achieves a Hit@10 of 38.34 across eight benchmarks, surpassing GPT-5.5. Furthermore, significant routing gains on validation sets confirm that structured knowledge injection substantially enhances the competitiveness of compact models in rare disease diagnosis.

Abstract

Rare-disease diagnosis is a long-tail reasoning problem: phenotypes are incomplete, individual disorders are sparsely documented, and relevant evidence is distributed across ontologies, gene annotations, and biomedical text. Language models consequently favor common conditions, miss rare candidates, or produce plausible but invalid names. We introduce RareDx, which couples controlled evidence use with knowledge-graph-grounded policy optimization. RareDx-Harness normalizes heterogeneous records into one ranked-diagnosis task and compares direct inference, static retrieval, adaptive tools, and structured phenotype-gene-disease reasoning over a shared knowledge layer. The training pipeline combines Top-10 post-training with RareDx-KGPO, our knowledge-graph-grounded policy optimization method. Its reward projects predictions into a canonical disease graph and integrates curated graded relevance, ontology proximity, biomedical similarity, and phenotype consistency. Vocabulary and output-budget constraints prevent dense partial credit from rewarding fabricated or overlong differentials. Across eight benchmarks, the complete RareDx system centered on Qwen3.5-9B reaches 38.34 macro Hit@10, 1.60 points above GPT-5.5 under the archived protocol; a disjoint validation-selection audit retains a 6.80-point routing gain over Direct on held-out cases. The 27B system reaches 23.53/36.56/40.76 at Hit@1/5/10. Controlled ablations show that retrieval is not uniformly helpful and that controlled routing is central to the gain. These results indicate that structured medical knowledge can turn a compact model into a competitive diagnostic ranker across heterogeneous long-tail settings in clinical practice.

Problem

• rare-disease diagnosis
• long-tail reasoning
• language models
• phenotype
• knowledge graph

Innovation

• Rare-Disease Diagnosis
• Knowledge Graph-Grounded Policy Optimization
• Controlled Knowledge Integration
• Long-Tail Reasoning
• Structured Phenotype-Gene-Disease Reasoning