MeddeX：结合RAG与Gemini Vision的LLaMA-3医疗报告分析系统
原标题：MeddeX: A Retrieval-Augmented LLaMA-3 System with Gemini Vision for Evidence-Backed Medical Report Analysis
AI 摘要
该论文提出 MeddeX，一个面向医疗报告分析的系统，将 QLoRA 微调的 LLaMA-3（8B）模型与检索增强生成（RAG）及 Gemini Vision 多模态层结合，用于生成准确、简洁且有证据支撑的医疗文档摘要。系统支持文本型 PDF、扫描图像 PDF 和放射影像，并引入垃圾文本检测、报告范围检索、纵向报告对比和惰性回填管线。与基础 LLaMA3-Med42-8B 模型相比，MeddeX 在 ROUGE-1、ROUGE-2、ROUGE-L、BLEU 和 BERTScore 上均有提升，验证了领域微调的效果。
正文节选
MeddeX: A Retrieval-Augmented LLaMA-3 System with Gemini Vision for Evidence-Backed Medical Report Analysis Abstract Modern healthcare generates increasingly complex medical reports that challenge clinical professionals in terms of time and interpretive burden. Manual review processes are slow, error-prone, and difficult to scale. This paper presents MeddeX, a Medical Report Analyzer that combines a QLoRA fine-tuned LLaMA-3 (8B) model with Retrieval-Augmented Generation (RAG) and a multimodal Ge