arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2603.22322cs.LGcs.AIcs.CY
收藏

AEGIS:一种用于美国和欧盟法规下适应性医疗AI市场后治理的操作基础设施

AEGIS: An Operational Infrastructure for Post-Market Governance of Adaptive Medical AI Under US and EU Regulations

Department of Clinical Science, Intervention and Technology, Karolinska Institutet(临床科学、干预与技术部门,Karolinska研究院);Department of Medical Radiation Physics, Stockholm University(医学辐射物理学部门,斯德哥尔摩大学);Department of Oncology-Pathology, Karolinska Institutet(肿瘤学-病理学部门,Karolinska研究院);Department of Clinical Physiology, Karolinska University Hospital(临床生理学部门,Karolinska大学医院);Department of Textile Technology, University of Bor s(纺织技术部门,Bor s大学);Department of Medical Technologies, Karolinska University Hospital(医学技术部门,Karolinska大学医院);Department of Biomedical Engineering and Health System, KTH Royal Institute of Technology(生物医学工程与健康系统部门,KTH皇家理工学院)

Fardin Afdideh, Mehdi Astaraki, Fernando Seoane, Farhad Abtahi

AI总结:

本文提出AEGIS框架,通过数据集整合、模型监控和条件决策模块,实现FDA PCCP和EU AI Act条款的执行,支持医疗AI的持续学习与安全更新。

摘要:

部署在医疗设备中的机器学习系统需要确保安全性的治理框架,以支持持续改进。FDA和欧洲联盟已引入如预定变更控制计划(PCCP)和市场后监督(PMS)等机制,以管理迭代模型更新而无需重复提交。本文提出了AI/ML评估和治理基础设施(AEGIS),适用于任何医疗AI系统。AEGIS包含三个模块:数据集整合与再训练、模型监控和条件决策,以执行FDA PCCP和EU AI Act第43(4)条的规定。我们实现了一个四类部署决策分类法(APPROVE、CONDITIONAL APPROVAL、CLINICAL REVIEW、REJECT)以及独立的PMS ALARM信号,能够在无可部署模型存在的情况下检测到关键状态,同时释放的模型同时处于风险状态。为了展示AEGIS如何在异构临床环境中实例化,我们提供了两个示例:从电子健康记录预测败血症和从医学影像进行脑肿瘤分割。两个案例使用相同的治理架构,仅在配置上不同。在11次模拟迭代中,AEGIS产生了8次APPROVE、1次CONDITIONAL APPROVAL、1次CLINICAL REVIEW和1次REJECT决策,行使了所有四类。ALARM信号在第8和第10次迭代中同时发布,包括在无可部署模型存在的情况下,释放的模型同时失效的临界状态。AEGIS在性能降级前检测到漂移。这些结果表明,AEGIS将监管变更控制概念转化为可执行的治理程序,支持在多样化临床应用中安全持续学习的适应性医疗AI。

英文摘要:

Machine learning systems deployed in medical devices require governance frameworks that ensure safety while enabling continuous improvement. Regulatory bodies including the FDA and European Union have introduced mechanisms such as the Predetermined Change Control Plan (PCCP) and Post-Market Surveillance (PMS) to manage iterative model updates without repeated submissions. This paper presents AI/ML Evaluation and Governance Infrastructure for Safety (AEGIS), a governance framework applicable to any healthcare AI system. AEGIS comprises three modules, i.e., dataset assimilation and retraining, model monitoring, and conditional decision, that operationalize FDA PCCP and EU AI Act Article 43(4) provisions. We implement a four-category deployment decision taxonomy (APPROVE, CONDITIONAL APPROVAL, CLINICAL REVIEW, REJECT) with an independent PMS ALARM signal, enabling detection of the critical state in which no deployable model exists while the released model is simultaneously at risk. To illustrate how AEGIS can be instantiated across heterogeneous clinical contexts, we provide two examples: sepsis prediction from electronic health records and brain tumor segmentation from medical imaging. Both cases use identical governance architecture, differing only in configuration. Across 11 simulated iterations on the sepsis example, AEGIS yielded 8 APPROVE, 1 CONDITIONAL APPROVAL, 1 CLINICAL REVIEW, and 1 REJECT decision, exercising all four categories. ALARM signals were co-issued at iterations 8 and 10, including the critical state where no deployable model exists and the released model is simultaneously failing. AEGIS detected drift before observable performance degradation. These results demonstrate that AEGIS translates regulatory change-control concepts into executable governance procedures, supporting safe continuous learning for adaptive medical AI across diverse clinical applications.

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。

↑