Algorithmic fairness in artificial intelligence for medicine and healthcare.

Algorithmic fairness in artificial intelligence for medicine and healthcare.
复制标题

DOI:
10.1038/s41551-023-01056-8
复制
发表时间:
2023-06
影响因子:
28.1
通讯作者:
--
中科院分区:
工程技术1区
文献类型:
--
作者:

文献摘要

参考文献

被引文献

相似文献

在医疗保健领域,开发和应用公平性不足的人工智能系统会影响公平医疗服务的提供。对按亚群体分层的人工智能模型的评估显示,在患者诊断、治疗以及医疗费用结算等方面存在不平等现象。在本《观点》文章中,我们从医疗保健的视角概述机器学习中的公平性问题,并探讨算法偏差(尤其是数据采集、基因变异和观察者内部标注差异方面的偏差)如何在临床工作流程中产生,以及可能导致的医疗保健差距。我们还审视了通过解缠、联邦学习和模型可解释性来减轻偏差的新兴技术,以及这些技术在基于人工智能的软件作为医疗器械开发过程中所发挥的作用。
In healthcare, the development and deployment of insufficiently fair systems of artificial intelligence can undermine the delivery of equitable care. Assessments of AI models stratified across sub-populations have revealed inequalities in how patients are diagnosed, given treatments, and billed for healthcare costs. In this Perspective, we outline fairness in machine learning through the lens of healthcare, and discuss how algorithmic biases (in data acquisition, genetic variation and intra-observer labelling variability, in particular) arise in clinical workflows and the healthcare disparities that they can cause. We also review emerging technology for mitigating biases via disentanglement, federated learning and model explainability, and their role in the development of AI-based software as a medical device.
DOI: 10.1038/s41551-022-00867-5
发表时间: 2022-12
影响因子: 28.1
作者:
Babenko B;Mitani A;Traynis I;Kitade N;Singh P;Maa AY;Cuadros J;Corrado GS;Peng L;Webster DR;Varadarajan A;Hammel N;Liu Y
通讯作者: Liu Y
DOI: 10.1001/jama.2017.14585
发表时间: 2017-12-12
影响因子: 120.7
作者:
Bejnordi, Babak Ehteshami;Veta, Mitko;van der Laak, Jeroen A. W. M.
通讯作者: van der Laak, Jeroen A. W. M.
DOI: 10.15779/z38bg31
发表时间: 2016-06-01
影响因子: 2.4
作者:
Barocas, Solon;Selbst, Andrew D.
通讯作者: Selbst, Andrew D.
DOI: 10.1158/1078-0432.ccr-19-2659
发表时间: 2020-04-01
影响因子: 11.5
作者:
Bhargava, Hersh K.;Leo, Patrick;Madabhushi, Anant
通讯作者: Madabhushi, Anant
DOI: 10.1038/s41586-018-0637-6
发表时间: 2018-11-01
期刊: NATURE
影响因子: 64.8
作者:
Awad, Edmond;Dsouza, Sohan;Rahwan, Iyad
通讯作者: Rahwan, Iyad