MeddeX: A Retrieval-Augmented LLaMA-3 System with Gemini Vision for Evidence-Backed Medical Report Analysis
Title: MeddeX: A Retrieval-Augmented LLaMA-3 System with Gemini Vision for Evidence-Backed Medical Report Analysis
标题：MeddeX：一个结合 Gemini Vision 的检索增强型 LLaMA-3 医学报告证据分析系统
Computer Science > Computation and Language arXiv:2610.20632 (cs) [Submitted on 30 Sep 2026] Title: MeddeX: A Retrieval-Augmented LLaMA-3 System with Gemini Vision for Evidence-Backed Medical Report Analysis Authors: Rajdeep Pandhare, Rohit Kundaikar, Yakub Kojar, Shreyansh Tiwari
计算机科学 > 计算与语言 arXiv:2610.20632 (cs) [提交于 2026 年 9 月 30 日] 标题：MeddeX：一个结合 Gemini Vision 的检索增强型 LLaMA-3 医学报告证据分析系统 作者：Rajdeep Pandhare, Rohit Kundaikar, Yakub Kojar, Shreyansh Tiwari
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 Gemini Vision layer to deliver accurate, concise, and evidence-backed summaries of medical documents.
摘要：现代医疗保健产生了日益复杂的医学报告，给临床专业人员带来了时间和解读负担方面的挑战。人工审查过程缓慢、容易出错且难以扩展。本文介绍了 MeddeX，这是一种医学报告分析器，它结合了经过 QLoRA 微调的 LLaMA-3 (8B) 模型、检索增强生成 (RAG) 和多模态 Gemini Vision 层，旨在提供准确、简洁且有证据支持的医学文档摘要。
The system handles text-native PDFs, scanned image-based PDFs, and direct radiological images through an intelligent hybrid extraction pipeline that includes a novel garbage-text detection mechanism for corrupted and UTF-16-encoded documents. Beyond single-report Q&A, MeddeX introduces report-scoped retrieval for precise contextual grounding, an automated longitudinal report comparison engine, and a self-healing backfill pipeline that lazily processes legacy documents on demand.
该系统通过智能混合提取流水线处理原生文本 PDF、基于扫描图像的 PDF 以及直接的放射影像，其中包括一种针对损坏和 UTF-16 编码文档的新型垃圾文本检测机制。除了单份报告的问答功能外，MeddeX 还引入了报告范围内的检索以实现精确的上下文基础，以及自动化的纵向报告比较引擎和按需处理旧文档的自愈式回填流水线。
Evaluation against the base LLaMA3-Med42-8B model demonstrates consistent improvements across ROUGE-1 (+9.2%), ROUGE-2 (+10.0%), ROUGE-L (+11.8%), BLEU (+71.4%), and BERTScore (+0.8%), validating the benefit of domain-specific fine-tuning. MeddeX represents a practical and robust step toward AI-assisted clinical decision support.
针对基础 LLaMA3-Med42-8B 模型的评估显示，ROUGE-1 (+9.2%)、ROUGE-2 (+10.0%)、ROUGE-L (+11.8%)、BLEU (+71.4%) 和 BERTScore (+0.8%) 均有持续提升，验证了特定领域微调的益处。MeddeX 代表了迈向人工智能辅助临床决策支持的实用且稳健的一步。