AI in Computational Pathology of Cancer: Improving Diagnostic Workflows and Clinical Outcomes?

IF 4.7 2区医学 Q1 ONCOLOGY

Annual Review of Cancer Biology-Series Pub Date : 2023-01-17 DOI:10.1146/annurev-cancerbio-061521-092038

D. Cifci, G. P. Veldhuizen, S. Foersch, Jakob Nikolas Kather

{"title":"AI in Computational Pathology of Cancer: Improving Diagnostic Workflows and Clinical Outcomes?","authors":"D. Cifci, G. P. Veldhuizen, S. Foersch, Jakob Nikolas Kather","doi":"10.1146/annurev-cancerbio-061521-092038","DOIUrl":null,"url":null,"abstract":"Histopathology plays a fundamental role in the diagnosis and subtyping of solid tumors and has become a cornerstone of modern precision oncology. Histopathological evaluation is typically performed manually by expert pathologists due to the complexity of visual data. However, in the last ten years, new artificial intelligence (AI) methods have made it possible to train computers to perform visual tasks with high performance, reaching similar levels as experts in some applications. In cancer histopathology, these AI tools could help automate repetitive tasks, making more efficient use of pathologists’ time. In research studies, AI methods have been shown to have an astounding ability to predict genetic alterations and identify prognostic and predictive biomarkers directly from routine tissue slides. Here, we give an overview of these recent applications of AI in computational pathology, focusing on new tools for cancer research that could be pivotal in identifying clinical biomarkers for better treatment decisions. Expected final online publication date for the Annual Review of Cancer Biology, Volume 7 is April 2023. Please see http://www.annualreviews.org/page/journal/pubdates for revised estimates.","PeriodicalId":54233,"journal":{"name":"Annual Review of Cancer Biology-Series","volume":" ","pages":""},"PeriodicalIF":4.7000,"publicationDate":"2023-01-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"5","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Annual Review of Cancer Biology-Series","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1146/annurev-cancerbio-061521-092038","RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"ONCOLOGY","Score":null,"Total":0}

引用次数: 5

Abstract

Histopathology plays a fundamental role in the diagnosis and subtyping of solid tumors and has become a cornerstone of modern precision oncology. Histopathological evaluation is typically performed manually by expert pathologists due to the complexity of visual data. However, in the last ten years, new artificial intelligence (AI) methods have made it possible to train computers to perform visual tasks with high performance, reaching similar levels as experts in some applications. In cancer histopathology, these AI tools could help automate repetitive tasks, making more efficient use of pathologists’ time. In research studies, AI methods have been shown to have an astounding ability to predict genetic alterations and identify prognostic and predictive biomarkers directly from routine tissue slides. Here, we give an overview of these recent applications of AI in computational pathology, focusing on new tools for cancer research that could be pivotal in identifying clinical biomarkers for better treatment decisions. Expected final online publication date for the Annual Review of Cancer Biology, Volume 7 is April 2023. Please see http://www.annualreviews.org/page/journal/pubdates for revised estimates.

查看原文本刊更多论文

人工智能在癌症计算病理学中的应用:改善诊断工作流程和临床结果?

组织病理学在实体瘤的诊断和分型中起着重要作用，已成为现代精确肿瘤学的基石。由于视觉数据的复杂性，组织病理学评估通常由专业病理学家手动进行。然而，在过去的十年里，新的人工智能（AI）方法使训练计算机执行高性能视觉任务成为可能，在某些应用中达到了与专家相似的水平。在癌症组织病理学中，这些人工智能工具可以帮助自动化重复任务，更有效地利用病理学家的时间。在研究中，人工智能方法已被证明具有惊人的能力，可以直接从常规组织切片中预测基因改变并识别预后和预测性生物标志物。在这里，我们概述了人工智能在计算病理学中的这些最新应用，重点关注癌症研究的新工具，这些工具可能在识别临床生物标志物以做出更好的治疗决策方面发挥关键作用。《癌症生物学年度评论》第7卷预计最终在线出版日期为2023年4月。请参阅http://www.annualreviews.org/page/journal/pubdates用于修订估算。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

Annual Review of Cancer Biology-Series Medicine-Oncology

CiteScore

14.50

自引率

1.30%

发文量

期刊介绍： The Annual Review of Cancer Biology offers comprehensive reviews on various topics within cancer research, covering pivotal and emerging areas in the field. As our understanding of cancer's fundamental mechanisms deepens and more findings transition into targeted clinical treatments, the journal is structured around three main themes: Cancer Cell Biology, Tumorigenesis and Cancer Progression, and Translational Cancer Science. The current volume of this journal has transitioned from gated to open access through Annual Reviews' Subscribe to Open program, ensuring all articles are published under a CC BY license.