基础模型时代

IF 81.1 1区医学 Q1 ONCOLOGY Nature Reviews Clinical Oncology Pub Date : 2024-09-05 DOI:10.1038/s41571-024-00941-8

Jana Lipkova, Jakob Nikolas Kather

{"title":"基础模型时代","authors":"Jana Lipkova, Jakob Nikolas Kather","doi":"10.1038/s41571-024-00941-8","DOIUrl":null,"url":null,"abstract":"The development of clinically relevant artificial intelligence (AI) models has traditionally required access to extensive labelled datasets, which inevitably centre AI advances around large centres and private corporations. Data availability has also dictated the development of AI applications: most studies focus on common cancer types, and leave rare diseases behind. However, this paradigm is changing with the advent of foundation models, which enable the training of more powerful and robust AI systems using much smaller datasets.","PeriodicalId":19079,"journal":{"name":"Nature Reviews Clinical Oncology","volume":"21 11","pages":"769-770"},"PeriodicalIF":81.1000,"publicationDate":"2024-09-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"The age of foundation models\",\"authors\":\"Jana Lipkova, Jakob Nikolas Kather\",\"doi\":\"10.1038/s41571-024-00941-8\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"The development of clinically relevant artificial intelligence (AI) models has traditionally required access to extensive labelled datasets, which inevitably centre AI advances around large centres and private corporations. Data availability has also dictated the development of AI applications: most studies focus on common cancer types, and leave rare diseases behind. However, this paradigm is changing with the advent of foundation models, which enable the training of more powerful and robust AI systems using much smaller datasets.\",\"PeriodicalId\":19079,\"journal\":{\"name\":\"Nature Reviews Clinical Oncology\",\"volume\":\"21 11\",\"pages\":\"769-770\"},\"PeriodicalIF\":81.1000,\"publicationDate\":\"2024-09-05\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Nature Reviews Clinical Oncology\",\"FirstCategoryId\":\"3\",\"ListUrlMain\":\"https://www.nature.com/articles/s41571-024-00941-8\",\"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 Clinical Oncology","FirstCategoryId":"3","ListUrlMain":"https://www.nature.com/articles/s41571-024-00941-8","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

传统上，开发临床相关的人工智能（AI）模型需要获取大量标记数据集，这就不可避免地使人工智能的发展集中在大型中心和私营企业周围。数据的可用性也决定了人工智能应用的发展：大多数研究都集中在常见癌症类型上，而将罕见疾病抛在脑后。然而，随着基础模型的出现，这种模式正在发生变化，它可以使用小得多的数据集来训练更强大、更稳健的人工智能系统。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

摘要图片

查看原文

微信好友朋友圈 QQ好友复制链接

本刊更多论文

The age of foundation models

The development of clinically relevant artificial intelligence (AI) models has traditionally required access to extensive labelled datasets, which inevitably centre AI advances around large centres and private corporations. Data availability has also dictated the development of AI applications: most studies focus on common cancer types, and leave rare diseases behind. However, this paradigm is changing with the advent of foundation models, which enable the training of more powerful and robust AI systems using much smaller datasets.

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 Clinical Oncology 医学-肿瘤学

CiteScore

99.40

自引率

0.40%

发文量

114

审稿时长

6-12 weeks

期刊介绍： Nature Reviews publishes clinical content authored by internationally renowned clinical academics and researchers, catering to readers in the medical sciences at postgraduate levels and beyond. Although targeted at practicing doctors, researchers, and academics within specific specialties, the aim is to ensure accessibility for readers across various medical disciplines. The journal features in-depth Reviews offering authoritative and current information, contextualizing topics within the history and development of a field. Perspectives, News & Views articles, and the Research Highlights section provide topical discussions, opinions, and filtered primary research from diverse medical journals.