End-to-End diagnosis of breast biopsy images with transformers.

End-to-End diagnosis of breast biopsy images with transformers.
复制标题

使用转换器实现乳腺活检图像的端到端诊断。

DOI:
10.1016/j.media.2022.102466
复制
发表时间:
2022-07
影响因子:
10.9
通讯作者:
Shapiro, Linda G.
Shapiro, Linda G.
中科院分区:
工程技术1区
文献类型:
--
作者:
Mehta, Sachin;Lu, Ximing;Wu, Wenjun;Weaver, Donald;Hajishirzi, Hannaneh;Elmore, Joann G.;Shapiro, Linda G.

文献摘要

参考文献

被引文献

相似文献

病理学家之间的诊断分歧存在于从良性病变到恶性病变的各个阶段。能够减少不确定性的计算机辅助诊断系统将对临床产生重要影响。为了开发一种计算机辅助诊断方法,将乳腺活检图像分类为一系列诊断类别(良性、非典型性、导管原位癌和浸润性乳腺癌),我们引入了一个基于转换器的全息注意网络HATNet。与最先进的组织病理图像分类系统使用双管齐下的方法不同,即它们首先使用多实例学习框架学习局部表示,然后结合这些局部表示来产生图像级别的决策,HATNet简化了组织病理图像分类流水线,并展示了如何端到端从十亿像素大小的图像学习表示。HATNet扩展了词袋方法,并使用自我注意来编码全局信息,使其能够从临床相关的组织结构中学习表示,而无需任何明确的监督。它的性能比以前最好的网络Y-Net高出8%,后者使用组织级分割掩码的形式进行监督。重要的是,我们的分析表明,HATNet从临床相关结构中学习表示,它与87名美国病理学家对这一具有挑战性的测试集的分类准确性相匹配。
Diagnostic disagreements among pathologists occur throughout the spectrum of benign to malignant lesions. A computer-aided diagnostic system capable of reducing uncertainties would have important clinical impact. To develop a computer-aided diagnosis method for classifying breast biopsy images into a range of diagnostic categories (benign, atypia, ductal carcinoma in situ, and invasive breast cancer), we introduce a transformer-based hollistic attention network called HATNet. Unlike state-of-the-art histopatho-logical image classification systems that use a two pronged approach, i.e., they first learn local representations using a multi-instance learning framework and then combine these local representations to produce image-level decisions, HATNet streamlines the histopathological image classification pipeline and shows how to learn representations from gigapixel size images end-to-end. HATNet extends the bag-of-words approach and uses self-attention to encode global information, allowing it to learn representations from clinically relevant tissue structures without any explicit supervision. It outperforms the previous best network Y-Net, which uses supervision in the form of tissue-level segmentation masks, by 8%. Importantly, our analysis reveals that HATNet learns representations from clinically relevant structures, and it matches the classification accuracy of 87 U.S. pathologists for this challenging test set.
DOI: 10.1016/j.semcdb.2009.10.003
发表时间: 2010-02
影响因子: 7.3
作者:
Arendt LM;Rudnick JA;Keller PJ;Kuperwasser C
通讯作者: Kuperwasser C
DOI: 10.1001/jama.2015.1405
发表时间: 2015-03-17
期刊: JAMA
影响因子: --
作者:
Elmore JG;Longton GM;Carney PA;Geller BM;Onega T;Tosteson AN;Nelson HD;Pepe MS;Allison KH;Schnitt SJ;O'Malley FP;Weaver DL
通讯作者: Weaver DL
DOI: 10.3322/caac.21583
发表时间: 2019-11-01
影响因子: 254.7
作者:
DeSantis, Carol E.;Ma, Jiemin;Siegel, Rebecca L.
通讯作者: Siegel, Rebecca L.
DOI: 10.1109/cvpr.2016.266
发表时间: 2016-06
期刊: Proceedings. IEEE Computer Society Conference on Computer Vision and Pattern Recognition
影响因子: --
作者:
Hou L;Samaras D;Kurc TM;Gao Y;Davis JE;Saltz JH
通讯作者: Saltz JH
DOI: 10.1111/his.12387
发表时间: 2014-08-01
期刊: HISTOPATHOLOGY
影响因子: 6.4
作者:
Allison, Kimberly H.;Reisch, Lisa M.;Elmore, Joann G.
通讯作者: Elmore, Joann G.