Large Language Model Aided Birt-Hogg-Dube Syndrome Diagnosis with Multimodal Retrieval-Augmented Generation
This work improves diagnostic accuracy for a rare disease (Birt-Hogg-Dubé syndrome) in clinical radiology, but it is incremental as it builds on existing multimodal large language models with domain-specific enhancements.
The paper tackled the problem of diagnosing Birt-Hogg-Dubé syndrome from CT images by addressing limited clinical samples and low inter-class differentiation in diffuse cystic lung diseases, proposing a multimodal retrieval-augmented generation framework that achieved superior accuracy and evidence-based descriptions aligned with expert insights.
Deep learning methods face dual challenges of limited clinical samples and low inter-class differentiation among Diffuse Cystic Lung Diseases (DCLDs) in advancing Birt-Hogg-Dube syndrome (BHD) diagnosis via Computed Tomography (CT) imaging. While Multimodal Large Language Models (MLLMs) demonstrate diagnostic potential fo such rare diseases, the absence of domain-specific knowledge and referable radiological features intensify hallucination risks. To address this problem, we propose BHD-RAG, a multimodal retrieval-augmented generation framework that integrates DCLD-specific expertise and clinical precedents with MLLMs to improve BHD diagnostic accuracy. BHDRAG employs: (1) a specialized agent generating imaging manifestation descriptions of CT images to construct a multimodal corpus of DCLDs cases. (2) a cosine similarity-based retriever pinpointing relevant imagedescription pairs for query images, and (3) an MLLM synthesizing retrieved evidence with imaging data for diagnosis. BHD-RAG is validated on the dataset involving four types of DCLDs, achieving superior accuracy and generating evidence-based descriptions closely aligned with expert insights.