AWCDL: Automatic weight calibration deep learning for detecting HER2 status in whole-slide breast cancer image

Shuai Wang; Xiaojing Guo; Jinlu Ma; Ning Wu; Xiaolong Qian; Guogui Sun; Jia Guo; Xiangfei Meng; Bo Kang; Bo Xu

doi:10.1016/j.intonc.2025.03.008

您当前的位置：

首页 >

文章列表页 >

AWCDL: Automatic weight calibration deep learning for detecting HER2 status in whole-slide breast cancer image

Research article | 更新时间：2026-06-03

- AWCDL: Automatic weight calibration deep learning for detecting HER2 status in whole-slide breast cancer image
- AWCDL: Automatic weight calibration deep learning for detecting HER2 status in whole-slide breast cancer image
- 智能肿瘤学 2025年1卷第2期页码：128-138
- 作者机构：
  
  1. Department of Radiation Oncology, Shandong Cancer Hospital and Institute, Shandong First Medical University and Shandong Academy of Medical Sciences,Shandong,Jinan,China,250117
  2. Department of Biochemistry and Molecular Biology, Key Laboratory of Breast Cancer Prevention and Therapy, Tianjin Clinical Research Center for Cancer, Ministry of Education, National Clinical Research Center for Cancer, Tianjin Medical University Cancer Institute & Hospital,Tianjin,China,300060
  3. Department of Breast Cancer Pathology and Research Laboratory, Tianjin Medical University Cancer Institute & Hospital,Tianjin,China,300060
  4. Department of Radiation Oncology, The First Affiliated Hospital of Xi’an Jiaotong University,Shaanxi,Xi'an,China,710061
  5. State Key Laboratory of Female Fertility Promotion, Center for Reproductive Medicine, Department of Obstetrics and Gynecology, Peking University Third Hospital,Beijing,China,100191
  6. Department of Hebei Key Laboratory of Medical–Industrial Integration Precision Medicine, Affiliated Hospital, North China University of Science and Technology,Hebei,Tangshan,China,063000
  7. Department of Research and Development, National Supercomputer Center in Tianjin,Tianjin,China,300000
  8. Chongqing Key Laboratory of Intelligent Oncology for Breast Cancer, Chongqing University Cancer Hospital, Chongqing University School of Medicine,Chongqing,China,400030
- 作者简介：
- 基金信息：
- DOI：10.1016/j.intonc.2025.03.008
  中图分类号：
- 收稿：2025-01-31，
  
  网络首发：2025-03-27，
  
  纸质出版：2025-04-24
- 稿件说明：
移动端阅览
Shuai Wang, Xiaojing Guo, Jinlu Ma, 等. AWCDL: Automatic weight calibration deep learning for detecting HER2 status in whole-slide breast cancer image[J]. 智能肿瘤学, 2025,1(2):128-138.

Shuai Wang, Xiaojing Guo, Jinlu Ma, et al. AWCDL: Automatic weight calibration deep learning for detecting HER2 status in whole-slide breast cancer image[J]. Intelligent Oncology, 2025, 1(2): 128-138.
Shuai Wang, Xiaojing Guo, Jinlu Ma, 等. AWCDL: Automatic weight calibration deep learning for detecting HER2 status in whole-slide breast cancer image[J]. 智能肿瘤学, 2025,1(2):128-138. DOI： 10.1016/j.intonc.2025.03.008.

Shuai Wang, Xiaojing Guo, Jinlu Ma, et al. AWCDL: Automatic weight calibration deep learning for detecting HER2 status in whole-slide breast cancer image[J]. Intelligent Oncology, 2025, 1(2): 128-138. DOI： 10.1016/j.intonc.2025.03.008.

摘要

Defining an

ERBB2

(

HER2/neu

) gene amplification status is critical to guiding human epidermal growth factor receptor 2 (HER2)-targeted therapy in breast cancer. Up to 40% of breast cancer patients are reported as having an immunohistochemistry (IHC) of HER2 2+ and requiring additional testing using fluorescence

in situ

hybridization to confirm the results. This paper aims to establish an automatically weighted calibration deep learning (AWCDL) algorithm to predict

ERBB2

amplification based on IHC images. In this study

we applied IHC HER2 2+ images from 1

073 breast cancer patients at three cancer centers in China and extracted 376

099 tiles. Among these

269

664 tiles were used for internal and external validation. The designed AWCDL consists of two steps. In Step 1

the internal validation achieved an accuracy of 89%

with a specificity of 0.89 and a sensitivity of 0.89. The external validation in the two other centers showed an average accuracy of 85%

with a specificity of 0.86 and a sensitivity of 0.82. In Ste

p 2

the model achieved higher accuracy for the slides predicted as negative in Step 1 by automatically calibrating the weight. Collectively

these results suggest that this AWCDL model has successfully proved useful as an alternative method to fluorescence

in situ

hybridization for assessing the

ERBB2

amplification status in breast cancer.

Abstract

Defining an

ERBB2

(

HER2/neu

in situ

hybridization to confirm the results. This paper aims to establish an automatically weighted calibration deep learning (AWCDL) algorithm to predict

ERBB2

amplification based on IHC images. In this study

we applied IHC HER2 2+ images from 1

073 breast cancer patients at three cancer centers in China and extracted 376

099 tiles. Among these

269

664 tiles were used for internal and external validation. The designed AWCDL consists of two steps. In Step 1

the internal validation achieved an accuracy of 89%

with a specificity of 0.89 and a sensitivity of 0.89. The external validation in the two other centers showed an average accuracy of 85%

with a specificity of 0.86 and a sensitivity of 0.82. In Step 2

the model achieved higher accuracy for the slides predicted as negative in Step 1 by automatically calibrating the weight. Collectively

these results suggest that this AWCDL model has successfully proved useful as an alternative method to fluorescence

in situ

hybridization for assessing the

ERBB2

amplification status in breast cancer.

关键词

Keywords

references

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

Multimodal medical imaging AI for breast cancer diagnosis: A comprehensive review

Analysis of key technologies and intelligent development trends for tumor surgery navigation platforms

Advancing precision oncology through hNQO1-activatable NIR-II probes: Integrating molecular imaging with artificial intelligence

Integrative multi-omics clustering for identifying novel breast cancer subtypes with distinct molecular and clinical characteristics

Integrating multi-omic liquid biopsies and artificial intelligence: The next frontier in early cancer detection