用本地大模型实现医疗表单自动填充，兼顾隐私与准确
sebis at CRF Filling 2026: A Two-Stage Local LLM Pipeline for Medical CRF Filling
- 分两阶段提取临床信息：先判断是否存在，再提取具体值
- 无需调用外部接口，仅用少量示例实现0.55的宏平均F1
- 适合注重数据安全的医疗机构使用
从非结构化电子病历中提取结构化临床信息是医疗信息学中的长期瓶颈。尽管大语言模型性能优异，但在临床场景部署时面临隐私风险、推理成本高以及幻觉问题。针对CL4Health 2026病例报告表（CRF）填写任务，我们提出一种完全本地化、领域适配的MedGemma-27B模型两阶段管道。该架构将存在性二分类与数值提取分离，严格遵循文本证据，确保否定、不确定或未知状态的确定性输出。通过项目特定的少量示例上下文学习，不依赖外部API调用或微调，模型在官方英文测试集上取得0.55的宏平均F1得分，位列所有本地部署开源方案第二。本工作表明，私密性保障的本地化大模型流程可达到接近商业前沿模型的性能，为临床自然语言处理提供可信赖、数据主权可控的解决方案。
原文摘要 · Abstract (English)
The extraction of structured clinical information from unstructured EHR notes is a persistent bottleneck in healthcare informatics. While large language models (LLMs) offer high performance, their deployment in clinical settings is hindered by privacy risks, inference costs, and the tendency to hallucinate beyond textual evidence. We address these challenges for the CL4Health 2026 Case Report Form (CRF) filling task by proposing a fully local, domain-adapted pipeline using the MedGemma-27B model. Our two-stage architecture, which separates binary presence classification from value extraction, enforces strict adherence to textual evidence and ensures deterministic outputs for negated, uncertain, or unknown states. By leveraging item-specific, few-shot in-context learning without external API calls or fine-tuning, our approach achieves a macro-F1 score of 0.55 on the official English test track. This result secures second place among all locally-hosted, open-source submissions. Our work demonstrates that privacy-preserving, on-premise LLM pipelines can achieve near-competitive performance with proprietary frontier models, providing a practical, data-sovereign framework for clinical NLP.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。