临床实验室检测监管对地方医学人工智能监管的启示-动脉网

Lessons for local oversight of AI in medicine from the regulation of clinical laboratory testing

Nature 等信源发布 2024-12-13 21:01



可切换为仅中文







Current regulatory frameworks for artificial intelligence-based clinical decision support (AICDS) are insufficient to ensure safety, effectiveness, and equity at the bedside. The oversight of clinical laboratory testing, which requires federal- and hospital-level involvement, offers many instructive lessons for how to balance safety and innovation and warnings regarding the fragility of this balance.

目前基于人工智能的临床决策支持（AICDS）的监管框架不足以确保床边的安全性，有效性和公平性。对临床实验室测试的监督需要联邦和医院层面的参与，这为如何平衡安全和创新提供了许多有益的教训，并警告了这种平衡的脆弱性。

We propose an AICDS oversight framework, modeled after clinical laboratory regulation, that is deliberative, inclusive, and collaborative..

我们提出了一个AICDS监督框架，该框架模仿临床实验室法规，具有审议性，包容性和协作性。。

With the growing availability of large electronic health record (EHR) datasets and powerful machine learning methods, the promise of effective and reliable artificial intelligence-based clinical decision support (AICDS) may be soon attainable. However, because of biases in EHR data, shifts in model performance over space and time, and the lack of evidence about the benefits and harms of AICDS in practice, this promise comes with considerable risk1,2,3,4.

随着大型电子健康记录（EHR）数据集和强大的机器学习方法的日益可用，有效可靠的基于人工智能的临床决策支持（AICDS）的前景可能很快就会实现。然而，由于EHR数据的偏差，模型性能随空间和时间的变化，以及缺乏关于AICD在实践中的益处和危害的证据，这一承诺带来了相当大的风险1,2,3,4。

As AICDS regulatory approaches continue to evolve, there is still no clear solution that balances safety and innovation and federal and local oversight. In this Comment, we review the oversight of clinical laboratory testing and discuss several lessons for developing novel regulatory approaches for AICDS to achieve these goals5.Oversight of AICDSIn the United States (US), although the Food and Drug Administration (FDA) and the Office of the National Coordinator for Health Information Technology (ONC) continue to develop regulatory frameworks for AICDS, these approaches alone will not guarantee the safety, effectiveness, and equity of all AICDS when deployed at the bedside6,7,8,9,10.

随着AICDS监管方法的不断发展，仍然没有明确的解决方案来平衡安全和创新以及联邦和地方监督。在这篇评论中，我们回顾了对临床实验室测试的监督，并讨论了为AICD开发新的监管方法以实现这些目标的几点教训5。对AICD的监督在美国（美国），尽管美国食品和药物管理局（FDA）和国家卫生信息技术协调员办公室（ONC）继续为AICD开发监管框架，但这些方法本身并不能保证所有AICD在床边部署时的安全性，有效性和公平性6,7,8,9,10。

First, many AICDS tools do not currently fall under the FDA or ONC purview, but still need oversight to ensure basic standards of quality and safety. Second, the predictive performance of many AICDS tools varies considerably in new settings and over time1,3. Thus, there is a need for a regulatory framework that ensures there is sufficient oversight of AICDS as it is deployed locally in individual hospitals and clinics.

首先，许多AICDS工具目前不属于FDA或ONC的职权范围，但仍需要监督以确保质量和安全的基本标准。其次，许多AICDS工具的预测性能在新设置和时间上差异很大1,3。因此，有必要建立一个监管框架，确保在个别医院和诊所部署AICD时对其进行充分监督。

Many health systems are working to figure out how to oversee AICDS, but these efforts are mostly at well-resourced academic centers and vary widely in their maturity11. Emerging medical AI networks9 are taking on important rol.

许多卫生系统正在研究如何监督AICDS，但这些努力大多在资源充足的学术中心进行，其成熟程度差异很大11。新兴的医疗AI网络9正在发挥重要作用。

ReferencesWong, A. et al. External validation of a widely implemented proprietary sepsis prediction model in hospitalized patients. JAMA Intern. Med. 181, 1065–1070 (2021).Article

ReferencesWong，A。等人。住院患者中广泛实施的专有脓毒症预测模型的外部验证。JAMA实习生。医学1811065-1070（2021）。文章

PubMed

Google Scholar

谷歌学者

Nastasi, A. J. et al. A vignette-based evaluation of ChatGPT’s ability to provide appropriate and equitable medical advice across care contexts. Sci. Rep. 13, 17885 (2023).Article

Nastasi，A.J.等人。基于小插曲的ChatGPT在整个护理环境中提供适当和公平医疗建议的能力评估。科学。代表1317885（2023）。文章

CAS

PubMed

PubMed Central

公共医学中心

Google Scholar

谷歌学者

Finlayson, S. G. et al. The clinician and dataset shift in artificial intelligence. New Engl. J. Med. 385, 283–286 (2021).Article

Finlayson，S.G.等人，《人工智能中的临床医生和数据集转变》。新英语。J、医学385283-286（2021）。文章

PubMed

Google Scholar

谷歌学者

Chen, S. et al. Use of artificial intelligence chatbots for cancer treatment information. JAMA Oncol. 9, 1459 (2023).Article

Chen，S.等人。使用人工智能聊天机器人获取癌症治疗信息。JAMA Oncol。91459（2023）。文章

PubMed

PubMed Central

PubMed 中心

Google Scholar

谷歌学者

Schulz, W. L., Durant, T. J. S. & Krumholz, H. M. Validation and regulation of clinical artificial intelligence. Clin. Chem. 65, 1336–1337 (2019).Article

Schulz，W.L.，Durant，T.J.S.＆Krumholz，H.M。临床人工智能的验证和监管。临床。化学。651336-1337（2019）。文章

CAS

PubMed

Google Scholar

谷歌学者

Price, W. N., Sendak, M., Balu, S. & Singh, K. Enabling collaborative governance of medical AI. Nat. Mach. Intell. 5, 821–823 (2023).Article

Price，W.N.，Sendak，M.，Balu，S。和Singh，K。实现医疗AI的协同治理。Nat。Mach。因特尔。5821-823（2023）。文章

Google Scholar

谷歌学者

Panch, T. et al. A distributed approach to the regulation of clinical AI. PLoS Digit. Health 1, e0000040 (2022).Article

Panch，T。等。临床AI的分布式调节方法。PLoS数字。健康1，e0000040（2022）。文章

PubMed

PubMed Central

PubMed 中心

Google Scholar

谷歌学者

Gerke, S., Babic, B., Evgeniou, T. & Cohen, I. G. The need for a system view to regulate artificial intelligence/machine learning-based software as medical device. NPJ Digit. Med. 3, 53 (2020).Article

Gerke，S.，Babic，B.，Evgeniou，T。＆Cohen，I.G。需要一个系统视图来规范人工智能/基于机器学习的软件作为医疗设备。NPJ数字。医学杂志3，53（2020）。文章

PubMed

PubMed Central

PubMed 中心

Google Scholar

谷歌学者

Shah, N. H. et al. A nationwide network of health AI Assurance laboratories. JAMA 331, 245 (2024).Article

Shah，N.H.等人。一个全国性的健康AI保证实验室网络。《美国医学会杂志》331245（2024）。文章

PubMed

Google Scholar

谷歌学者

Lee, J. T. et al. Analysis of devices authorized by the FDA for clinical decision support in critical care. JAMA Intern. Med. 183, 1399 (2023).Article

Lee，J.T.等人。FDA授权用于重症监护临床决策支持的设备分析。JAMA实习生。医学1831399（2023）。文章

PubMed

PubMed Central

PubMed 中心

Google Scholar

谷歌学者

Nong, P., Hamasha, R., Singh, K., Adler-Milstein, J. & Platt, J. How academic medical centers govern AI prediction tools in the context of uncertainty and evolving regulation. NEJM AI 1, AIp2300048 (2024).Article

Nong，P.，Hamasha，R.，Singh，K.，Adler-Milstein，J。＆Platt，J。学术医学中心如何在不确定性和不断变化的监管背景下管理AI预测工具。NEJM AI 1，AIp2300048（2024）。文章

Google Scholar

谷歌学者

Longhurst, C. A., Singh, K., Chopra, A., Atreja, A. & Brownstein, J. S. A call for artificial intelligence implementation science centers to evaluate clinical effectiveness. NEJM AI 1, AIp2400223 (2024).Article

Longhurst，C.A.，Singh，K.，Chopra，A.，Atreja，A。＆Brownstein，J.S。呼吁人工智能实施科学中心评估临床有效性。NEJM AI 1，AIp2400223（2024）。文章

Google Scholar

谷歌学者

Graden, K. C., Bennett, S. A., Delaney, S. R., Gill, H. E. & Willrich, M. A. V. A high-level overview of the regulations surrounding a clinical laboratory and upcoming regulatory challenges for laboratory developed tests. Lab. Med. 52, 315–328 (2021).Article

Graden，K.C.，Bennett，S.A.，Delaney，S.R.，Gill，H.E。＆Willrich，M.A.V。对围绕临床实验室的法规以及即将到来的实验室开发测试监管挑战的高水平概述。实验室医学52315-328（2021）。文章

PubMed

Google Scholar

谷歌学者

Genzen, J. R. Regulation of laboratory-developed tests. Am. J. Clin. Pathol. 152, 122–131 (2019).Article

Genzen，J.R。实验室开发测试的规定。美国J.克林。病理学。152122-131（2019）。文章

PubMed

PubMed Central

公共医学中心

Google Scholar

谷歌学者

Price, W. N., Gerke, S. & Cohen, I. G. Potential liability for physicians using artificial intelligence. JAMA 322, 1765 (2019).Article

Price，W.N.，Gerke，S。＆Cohen，I.G。使用人工智能的医生的潜在责任。JAMA 3221765（2019）。文章

PubMed

Google Scholar

谷歌学者

Paranjape, K. et al. The value of artificial intelligence in laboratory medicine. Am. J. Clin. Pathol. 155, 823–831 (2021).Article

Paranjape，K.等人，《人工智能在检验医学中的价值》。美国J.克林。病理学。155823-831（2021）。文章

PubMed

Google Scholar

谷歌学者

Bellini, C., Padoan, A., Carobene, A. & Guerranti, R. A survey on artificial intelligence and big data utilisation in Italian clinical laboratories. Clin. Chem. Lab. Med. (CCLM) 60, 2017–2026 (2022).Article

Bellini，C.，Padoan，A.，Carobene，A。和Guerranti，R。意大利临床实验室人工智能和大数据利用调查。临床。化学。实验室医学（CCLM）602017–2026（2022）。文章

CAS

PubMed

Google Scholar

谷歌学者

Mazer, B. Theranos exploited black box medicine. BMJ 379, o3003 (2022).Article

Mazer，B.Theranos开发了黑匣子药物。BMJ 379，o3003（2022）。文章

PubMed

Google Scholar

谷歌学者

Food & Drug Administration. Genetic Non-Invasive Prenatal Screening Tests May Have False Results: FDA Safety Communication. https://www.fda.gov/medical-devices/safety-communications/genetic-non-invasive-prenatal-screening-tests-may-have-false-results-fda-safety-communication (2022).Department of Health and Human Services, FDA.

食品和药物管理局。基因非侵入性产前筛查测试可能会产生错误结果：FDA安全通讯。https://www.fda.gov/medical-devices/safety-communications/genetic-non-invasive-prenatal-screening-tests-may-have-false-results-fda-safety-communication（2022年）。美国食品和药物管理局卫生与公众服务部。

Medical Devices; Laboratory Developed Tests. 21 CFR Part 809 [Docket No. FDA-2023-N-2177]. https://www.federalregister.gov/documents/2024/05/06/2024-08935/medical-devices-laboratory-developed-tests (2024).French, D. Clinical utility of laboratory developed mass spectrometry assays for steroid hormone testing.

医疗器械；实验室开发的测试。21 CFR第809部分【卷宗号FDA-2023-N-2177】。https://www.federalregister.gov/documents/2024/05/06/2024-08935/medical-devices-laboratory-developed-tests（2024年）。French，D。实验室开发的用于类固醇激素测试的质谱分析的临床实用性。

J. Mass Spectrom. Adv. Clin. Lab 28, 13–19 (2023).Article .

J、质谱。临床顾问。实验室28，13-19（2023）。文章。

CAS

PubMed

PubMed Central

公共医学中心

Google Scholar

谷歌学者

Marzinke, M. A. et al. The VALIDity of laboratory developed tests: leave it to the experts? J. Mass Spectrom. Adv. Clin. Lab 27, 1–6 (2023).Article

Marzinke，M.A.等人。实验室开发测试的有效性：留给专家吗？J、质谱。临床顾问。实验室27，1-6（2023）。文章

CAS

PubMed

Google Scholar

谷歌学者

Prenosis Sepsis ImmunoScore. Software Device To Aid In The Prediction Or Diagnosis Of Sepsis. Regulation Number 880.6316. https://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfpmn/denovo.cfm?id=DEN230036 (2024).Fleuren, L. M. et al. Machine learning for the prediction of sepsis: a systematic review and meta-analysis of diagnostic test accuracy.

产前败血症免疫评分。有助于预测或诊断败血症的软件设备。法规编号880.6316。https://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfpmn/denovo.cfm?id=DEN230036（2024年）。Fleuren，L.M.等人。用于预测败血症的机器学习：诊断测试准确性的系统综述和荟萃分析。

Intensive Care Med. 46, 383–400 (2020).Article .

重症监护医学46383-400（2020）。文章。

PubMed

PubMed Central

公共医学中心

Google Scholar

谷歌学者

Fleisher, L. A. & Economou-Zavlanos, N. J. Artificial intelligence can be regulated using current patient safety procedures and infrastructure in hospitals. JAMA Health Forum 5, e241369 (2024).Article

Fleisher，L.A。和Economou Zavlanos，N.J。可以使用医院当前的患者安全程序和基础设施来管理人工智能。JAMA健康论坛5，e241369（2024）。文章

PubMed

Google Scholar

谷歌学者

Download referencesAuthor informationAuthors and AffiliationsDepartment of Pathology & Laboratory Medicine, University of Pennsylvania, Philadelphia, PA, USADaniel S. Herman & Jenna T. ReeceCenter for AI and Data Science for Integrated Diagnostics, University of Pennsylvania, Philadelphia, PA, USADaniel S.

下载参考文献作者信息作者和附属机构美国宾夕法尼亚州费城宾夕法尼亚大学病理学与检验医学系Daniel S.Herman＆Jenna T.Reecenter for AI and Data Science for Integrated Diagnostics，宾夕法尼亚大学费城分校，宾夕法尼亚州，美国Daniel S。

Herman & Gary E. WeissmanLeonard Davis Institute of Health Economics, University of Pennsylvania, Philadelphia, PA, USADaniel S. Herman & Gary E. WeissmanDepartment of Medicine, University of Pennsylvania, Philadelphia, PA, USAGary E. WeissmanAuthorsDaniel S. HermanView author publicationsYou can also search for this author in.

Herman＆Gary E.WeissmanLeonard Davis美国宾夕法尼亚州费城宾夕法尼亚大学健康经济研究所Daniel S.Herman＆Gary E.WeissmanDepartment of Medicine，宾夕法尼亚州费城宾夕法尼亚大学，USAGary E.WeissmanAuthorsDaniel S.HermanView作者出版物您也可以在中搜索这位作者。

PubMed Google ScholarJenna T. ReeceView author publicationsYou can also search for this author in

PubMed Google ScholarJenna T.ReeceView作者出版物您也可以在

PubMed Google ScholarGary E. WeissmanView author publicationsYou can also search for this author in

PubMed Google ScholarGary E.WeissmanView作者出版物您也可以在

PubMed Google ScholarContributionsD.S.H. and G.E.W. conceived the idea of the manuscript. D.S.H., J.T.R., and G.E.W. wrote and reviewed the manuscript.Corresponding authorCorrespondence to

PubMed谷歌学术贡献SD。S、 H.和G.E.W.构思了手稿的想法。D、 S.H.，J.T.R。和G.E.W.撰写并审阅了手稿。对应作者对应

Daniel S. Herman.Ethics declarations

丹尼尔·S·赫尔曼。道德宣言

Competing interests

相互竞争的利益

The authors declare no competing interests.

作者声明没有利益冲突。

Additional informationPublisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.Rights and permissions

Additional informationPublisher的注释Springer Nature在已发布的地图和机构隶属关系中的管辖权主张方面保持中立。权限和权限

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material.

开放获取本文是根据知识共享署名非商业性NoDerivatives 4.0国际许可证授权的，该许可证允许以任何媒介或格式进行任何非商业性使用，共享，分发和复制，只要您对原始作者和来源给予适当的信任，提供知识共享许可证的链接，并指出您是否修改了许可材料。

You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/..

要查看此许可证的副本，请访问http://creativecommons.org/licenses/by-nc-nd/4.0/..

Reprints and permissionsAbout this articleCite this articleHerman, D.S., Reece, J.T. & Weissman, G.E. Lessons for local oversight of AI in medicine from the regulation of clinical laboratory testing.

转载和许可本文引用本文Herman，D.S.，Reece，J.T。＆Weissman，G.E。从临床实验室测试法规中对医学人工智能进行地方监督的经验教训。

npj Digit. Med. 7, 359 (2024). https://doi.org/10.1038/s41746-024-01369-1Download citationReceived: 14 September 2024Accepted: 03 December 2024Published: 13 December 2024DOI: https://doi.org/10.1038/s41746-024-01369-1Share this articleAnyone you share the following link with will be able to read this content:Get shareable linkSorry, a shareable link is not currently available for this article.Copy to clipboard.

npj数字。医学杂志7359（2024）。https://doi.org/10.1038/s41746-024-01369-1Download引文接收日期：2024年9月14日接受日期：2024年12月3日发布日期：2024年12月13日OI：https://doi.org/10.1038/s41746-024-01369-1Share本文与您共享以下链接的任何人都可以阅读此内容：获取可共享链接对不起，本文目前没有可共享的链接。复制到剪贴板。

Provided by the Springer Nature SharedIt content-sharing initiative

由Springer Nature SharedIt内容共享计划提供

全球产业链接平台

重庆市渝北区金星科技大厦A区5楼512室

联系电话：023-67139735（重庆）

关于我们

产品服务