从安全文档到安全知识支持：面向医疗器械的基于证据的大语言模型框架
From Safety Documentation to Safety Knowledge Support: An Evidence-Grounded LLM Framework for Medical Devices
浏览论文内容
中文总结 AI 辅助
针对现有LLM在医疗器械安全工程中源链接等支持不足的问题，本文提出基于证据的LLM框架，用于安全知识支持，不做安全判定，还给出对应评估策略。
中文摘要 AI 辅助
医疗器械正变得越来越依赖软件、互联互通且具备AI能力，其开发需符合ISO 14971标准的风险管理证据，软件领域还需符合IEC 62304标准，该证据需在需求、设计决策、软件变更、验证结果、投诉及上市后数据间保持一致。这些任务成本高昂，且依赖稀缺的安全与领域专家。大语言模型（LLM）或可减少部分工作量，因为医疗器械安全工作高度依赖文档。但当前基于LLM的安全工程研究常聚焦孤立方法，依赖通用提示或公开示例，在源链接、可追溯性、不确定性处理、生命周期更新及记录专家评审等方面支持有限，限制了其在受监管医疗器械开发中的应用。本文指出核心研究问题并非安全文本生成，而是源链接安全知识支持，提出一种基于证据的框架，该框架连接器械制品、受控知识存储与检索、特定方法的候选安全项生成、评审与不确定性检查及记录专家评审，为专家决策准备、链接、检查及更新候选安全制品，不判定器械是否安全，也不提供监管批准。本文还概述了评估策略，采用非公开或新建的医疗器械案例研究及专家参考分析，评估覆盖范围、正确性、相关性、可追溯性、重复率、无依据主张及评审工作量。
英文摘要
Medical devices are becoming more software-intensive, connected, and AI-enabled. Their development requires risk-management evidence aligned with ISO 14971 and, for software, IEC 62304. This evidence must be kept consistent across requirements, design decisions, software changes, verification results, complaints, and post-market data. These tasks are costly and depend on scarce safety and domain experts. Large language models (LLMs) may reduce parts of this effort because medical-device safety work is highly document-based. However, current LLM-based safety-engineering studies often address isolated methods, rely on generic prompting or public examples, and provide limited support for source links, traceability, uncertainty handling, lifecycle updates, and recorded expert review. This limits their use in regulated medical-device development. This paper argues that the central research problem is not safety-text generation, but source-linked safety-knowledge support. We propose an evidence-grounded framework that connects device artifacts, controlled knowledge storage and retrieval, method-specific generation of candidate safety items, critique and uncertainty checks, and recorded expert review. The framework prepares, links, checks, and updates candidate safety artifacts for expert decision-making. It does not decide whether a device is safe and does not provide regulatory approval. We also outline an evaluation strategy using non-public or newly built medical-device case studies and expert reference analyses to assess coverage, correctness, relevance, traceability, duplicate rate, unsupported claims, and review effort.
发表机构
- Fraunhofer IESE（弗劳恩霍夫IESE研究所）
- RPTU Kaiserslautern-Landau（凯撒斯劳滕-兰道莱茵-普法尔茨理工大学）
机构由 AI 辅助整理，请以论文原文为准。
