限制人工智能模型中的偏差，改善癌症护理的公平性。

IF 72.5 1区医学 Q1 ONCOLOGY

Nature Reviews Cancer Pub Date : 2024-08-27 DOI:10.1038/s41568-024-00739-x

Marzyeh Ghassemi, Alexander Gusev

{"title":"限制人工智能模型中的偏差，改善癌症护理的公平性。","authors":"Marzyeh Ghassemi, Alexander Gusev","doi":"10.1038/s41568-024-00739-x","DOIUrl":null,"url":null,"abstract":"Cancer screening, diagnosis and care stand to benefit greatly from advances in artificial intelligence (AI). Researchers, developers and deployers must ensure that applications of AI avoid known racial and gender biases to advance health care for all. Cancer screening, diagnosis and care can benefit greatly from advances in artificial intelligence (AI). In this Comment, Ghassemi and Gusev discuss how AI applications must address and avoid known racial and gender biases to improve health care for all.","PeriodicalId":19055,"journal":{"name":"Nature Reviews Cancer","volume":"24 12","pages":"823-824"},"PeriodicalIF":72.5000,"publicationDate":"2024-08-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Limiting bias in AI models for improved and equitable cancer care\",\"authors\":\"Marzyeh Ghassemi, Alexander Gusev\",\"doi\":\"10.1038/s41568-024-00739-x\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Cancer screening, diagnosis and care stand to benefit greatly from advances in artificial intelligence (AI). Researchers, developers and deployers must ensure that applications of AI avoid known racial and gender biases to advance health care for all. Cancer screening, diagnosis and care can benefit greatly from advances in artificial intelligence (AI). In this Comment, Ghassemi and Gusev discuss how AI applications must address and avoid known racial and gender biases to improve health care for all.\",\"PeriodicalId\":19055,\"journal\":{\"name\":\"Nature Reviews Cancer\",\"volume\":\"24 12\",\"pages\":\"823-824\"},\"PeriodicalIF\":72.5000,\"publicationDate\":\"2024-08-27\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Nature Reviews Cancer\",\"FirstCategoryId\":\"3\",\"ListUrlMain\":\"https://www.nature.com/articles/s41568-024-00739-x\",\"RegionNum\":1,\"RegionCategory\":\"医学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q1\",\"JCRName\":\"ONCOLOGY\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Nature Reviews Cancer","FirstCategoryId":"3","ListUrlMain":"https://www.nature.com/articles/s41568-024-00739-x","RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"ONCOLOGY","Score":null,"Total":0}

引用次数: 0

摘要

人工智能（AI）的进步将使癌症筛查、诊断和治疗受益匪浅。研究人员、开发人员和部署人员必须确保人工智能的应用避免已知的种族和性别偏见，以促进全民医疗保健。癌症筛查、诊断和治疗将从人工智能（AI）的进步中受益匪浅。在这篇评论中，Ghassemi 和 Gusev 讨论了人工智能应用必须如何解决和避免已知的种族和性别偏见，以改善所有人的医疗保健。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

查看原文本刊更多论文

Limiting bias in AI models for improved and equitable cancer care

Cancer screening, diagnosis and care stand to benefit greatly from advances in artificial intelligence (AI). Researchers, developers and deployers must ensure that applications of AI avoid known racial and gender biases to advance health care for all. Cancer screening, diagnosis and care can benefit greatly from advances in artificial intelligence (AI). In this Comment, Ghassemi and Gusev discuss how AI applications must address and avoid known racial and gender biases to improve health care for all.

求助全文

通过发布文献求助，成功后即可免费获取论文全文。去求助

来源期刊

Nature Reviews Cancer 医学-肿瘤学

CiteScore

111.90

自引率

0.40%

发文量

审稿时长

6-12 weeks

期刊介绍： Nature Reviews Cancer, a part of the Nature Reviews portfolio of journals, aims to be the premier source of reviews and commentaries for the scientific communities it serves. The correct abbreviation for abstracting and indexing purposes is Nat. Rev. Cancer. The international standard serial numbers (ISSN) for Nature Reviews Cancer are 1474-175X (print) and 1474-1768 (online). Unlike other journals, Nature Reviews Cancer does not have an external editorial board. Instead, all editorial decisions are made by a team of full-time professional editors who are PhD-level scientists. The journal publishes Research Highlights, Comments, Reviews, and Perspectives relevant to cancer researchers, ensuring that the articles reach the widest possible audience due to their broad scope.