arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2511.04720cs.CLcs.AI
收藏

通过检索增强智能体学习对罕见疾病进行推理

Learning to reason about rare diseases through retrieval-augmented agents

Technical University of Munich(慕尼黑技术大学);GE HealthCare(GE医疗);Munich Center for Machine Learning(慕尼黑机器学习中心);Klinikum Rechts der Isar(莱布尼茨医院);Helmholtz Munich(亥姆霍兹慕尼黑研究中心);King's College London(伦敦国王学院)

Ha Young Kim, Jun Li, Ana Beatriz Solana, Carolin M. Pirkl, Benedikt Wiestler, Julia A. Schnabel, Cosmin I. Bercea

AI总结:

本文提出RADAR检索增强诊断推理智能体,通过FAISS索引外部病例报告和文献,为脑MRI罕见病检测提供无需额外训练的模型无关推理模块,在NOVA数据集上实现最高10.2%的性能提升。

摘要:

罕见疾病代表医学影像的长尾分布,由于代表性训练数据稀缺,AI模型在这些疾病上往往表现不佳。在临床工作流程中,放射科医生面对不熟悉的影像发现时,经常查阅病例报告和文献。遵循这一推理思路,我们提出RADAR,即检索增强诊断推理智能体(Retrieval Augmented Diagnostic Reasoning Agents),这是一个用于脑MRI罕见疾病检测的智能体系统。我们的方法使用AI智能体访问外部医学知识,通过句子变换器(sentence transformers)对病例报告和文献进行嵌入,并使用FAISS建立索引,以实现高效的相似性搜索。该智能体检索临床相关证据,在无需额外训练的情况下指导对未见疾病的诊断决策。RADAR被设计为一个与模型无关的推理模块,可以无缝集成到多种大型语言模型中,持续提升它们对罕见病理的识别能力和可解释性。在包含280种不同罕见疾病的NOVA数据集上,RADAR实现了最高10.2%的性能提升,其中开源模型(如DeepSeek)的改进最为显著。除准确率之外,检索到的示例还提供了可解释的、有文献依据的解释,凸显了检索增强推理作为医学影像中低患病率疾病处理的一种强大范式。

英文摘要:

Rare diseases represent the long tail of medical imaging, where AI models often fail due to the scarcity of representative training data. In clinical workflows, radiologists frequently consult case reports and literature when confronted with unfamiliar findings. Following this line of reasoning, we introduce RADAR, Retrieval Augmented Diagnostic Reasoning Agents, an agentic system for rare disease detection in brain MRI. Our approach uses AI agents with access to external medical knowledge by embedding both case reports and literature using sentence transformers and indexing them with FAISS to enable efficient similarity search. The agent retrieves clinically relevant evidence to guide diagnostic decision making on unseen diseases, without the need of additional training. Designed as a model-agnostic reasoning module, RADAR can be seamlessly integrated with diverse large language models, consistently improving their rare pathology recognition and interpretability. On the NOVA dataset comprising 280 distinct rare diseases, RADAR achieves up to a 10.2% performance gain, with the strongest improvements observed for open source models such as DeepSeek. Beyond accuracy, the retrieved examples provide interpretable, literature grounded explanations, highlighting retrieval-augmented reasoning as a powerful paradigm for low-prevalence conditions in medical imaging.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑