An assessment of breast cancer HER2, ER, and PR expressions based on mammography using deep learning with convolutional neural networks
Abstract Mammography is the recommended imaging modality for breast cancer screening. Expressions of human epidermal growth factor receptor 2 (HER2), estrogen receptor (ER), and progesterone receptor (PR) are critical to the development of therapeutic strategies for breast cancer. In this study, a d...
Saved in:
| Main Authors: | , , , , , , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2025-02-01
|
| Series: | Scientific Reports |
| Subjects: | |
| Online Access: | https://doi.org/10.1038/s41598-024-83597-9 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1823862701018841088 |
|---|---|
| author | Shun Zeng Hongyu Chen Rui Jing Wenzhuo Yang Ligong He Tianle Zou Peng Liu Bo Liang Dan Shi Wenhao Wu Qiusheng Lin Zhenyu Ma Jinhui Zha Yonghao Zhong Xianbin Zhang Guangrui Shao Peng Gong |
| author_facet | Shun Zeng Hongyu Chen Rui Jing Wenzhuo Yang Ligong He Tianle Zou Peng Liu Bo Liang Dan Shi Wenhao Wu Qiusheng Lin Zhenyu Ma Jinhui Zha Yonghao Zhong Xianbin Zhang Guangrui Shao Peng Gong |
| author_sort | Shun Zeng |
| collection | DOAJ |
| description | Abstract Mammography is the recommended imaging modality for breast cancer screening. Expressions of human epidermal growth factor receptor 2 (HER2), estrogen receptor (ER), and progesterone receptor (PR) are critical to the development of therapeutic strategies for breast cancer. In this study, a deep learning model (CBAM ResNet-18) was developed to predict the expression of these three receptors on mammography without manual segmentation of masses. Mammography of patients with pathologically proven breast cancer was obtained from two centers. A deep learning-based model (CBAM ResNet-18) for predicting HER2, ER, and PR expressions was trained and validated using five-fold cross-validation on a training dataset. The performance of the model was further tested using an external test dataset. Area under receiver operating characteristic curve (AUC), accuracy (ACC), and F1-score were calculated to assess the ability of the model to predict each receptor. For comparison we also developed original ResNet-18 without attention module and VGG-19 with and without attention module. The AUC (95% CI), ACC, and F1-score were 0.708 (0.609, 0.808), 0.651, 0.528, respectively, in the HER2 test dataset; 0.785 (0.673, 0.897), 0.845, 0.905, respectively, in the ER test dataset; and 0.706 (0.603, 0.809), 0.678, 0.773, respectively, in the PR test dataset. The proposed model demonstrates superior performance compared to the original ResNet-18 without attention module and VGG-19 with and without attention module. The model has the potential to predict HER2, PR, and especially ER expressions, and thus serve as an adjunctive diagnostic tool for breast cancer. |
| format | Article |
| id | doaj-art-f010cf05eec74f37a991e2da7dde6d13 |
| institution | Kabale University |
| issn | 2045-2322 |
| language | English |
| publishDate | 2025-02-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Scientific Reports |
| spelling | doaj-art-f010cf05eec74f37a991e2da7dde6d132025-02-09T12:28:16ZengNature PortfolioScientific Reports2045-23222025-02-0115111010.1038/s41598-024-83597-9An assessment of breast cancer HER2, ER, and PR expressions based on mammography using deep learning with convolutional neural networksShun Zeng0Hongyu Chen1Rui Jing2Wenzhuo Yang3Ligong He4Tianle Zou5Peng Liu6Bo Liang7Dan Shi8Wenhao Wu9Qiusheng Lin10Zhenyu Ma11Jinhui Zha12Yonghao Zhong13Xianbin Zhang14Guangrui Shao15Peng Gong16Department of General Surgery, Institute of Precision Diagnosis and Treatment of Digestive System Tumors and Guangdong Provincial Key Laboratory of Chinese Medicine Ingredients and Gut Microbiomics, Shenzhen University General Hospital, Shenzhen UniversityDepartment of Health Outcomes & Biomedical Informatics, University of FloridaDepartment of Radiology, Second Hospital of Shandong UniversitySun Yat-sen University Cancer Center, Sun Yat-sen UniversitySun Yat-sen University Cancer Center, Sun Yat-sen UniversityDepartment of General Surgery, Institute of Precision Diagnosis and Treatment of Digestive System Tumors and Guangdong Provincial Key Laboratory of Chinese Medicine Ingredients and Gut Microbiomics, Shenzhen University General Hospital, Shenzhen UniversityDepartment of General Surgery, Institute of Precision Diagnosis and Treatment of Digestive System Tumors and Guangdong Provincial Key Laboratory of Chinese Medicine Ingredients and Gut Microbiomics, Shenzhen University General Hospital, Shenzhen UniversityDepartment of General Surgery, Institute of Precision Diagnosis and Treatment of Digestive System Tumors and Guangdong Provincial Key Laboratory of Chinese Medicine Ingredients and Gut Microbiomics, Shenzhen University General Hospital, Shenzhen UniversityDepartment of General Surgery, Institute of Precision Diagnosis and Treatment of Digestive System Tumors and Guangdong Provincial Key Laboratory of Chinese Medicine Ingredients and Gut Microbiomics, Shenzhen University General Hospital, Shenzhen UniversityDepartment of General Surgery, Institute of Precision Diagnosis and Treatment of Digestive System Tumors and Guangdong Provincial Key Laboratory of Chinese Medicine Ingredients and Gut Microbiomics, Shenzhen University General Hospital, Shenzhen UniversityDepartment of Thyroid and Breast Surgery, Huazhong University of Science and Technology Union Shenzhen HospitalDepartment of Radiology, Second Hospital of Shandong University Zhaoyuan BranchDepartment of General Surgery, Institute of Precision Diagnosis and Treatment of Digestive System Tumors and Guangdong Provincial Key Laboratory of Chinese Medicine Ingredients and Gut Microbiomics, Shenzhen University General Hospital, Shenzhen UniversityDepartment of General Surgery, Institute of Precision Diagnosis and Treatment of Digestive System Tumors and Guangdong Provincial Key Laboratory of Chinese Medicine Ingredients and Gut Microbiomics, Shenzhen University General Hospital, Shenzhen UniversityDepartment of General Surgery, Institute of Precision Diagnosis and Treatment of Digestive System Tumors and Guangdong Provincial Key Laboratory of Chinese Medicine Ingredients and Gut Microbiomics, Shenzhen University General Hospital, Shenzhen UniversityDepartment of Radiology, Second Hospital of Shandong UniversityDepartment of General Surgery, Institute of Precision Diagnosis and Treatment of Digestive System Tumors and Guangdong Provincial Key Laboratory of Chinese Medicine Ingredients and Gut Microbiomics, Shenzhen University General Hospital, Shenzhen UniversityAbstract Mammography is the recommended imaging modality for breast cancer screening. Expressions of human epidermal growth factor receptor 2 (HER2), estrogen receptor (ER), and progesterone receptor (PR) are critical to the development of therapeutic strategies for breast cancer. In this study, a deep learning model (CBAM ResNet-18) was developed to predict the expression of these three receptors on mammography without manual segmentation of masses. Mammography of patients with pathologically proven breast cancer was obtained from two centers. A deep learning-based model (CBAM ResNet-18) for predicting HER2, ER, and PR expressions was trained and validated using five-fold cross-validation on a training dataset. The performance of the model was further tested using an external test dataset. Area under receiver operating characteristic curve (AUC), accuracy (ACC), and F1-score were calculated to assess the ability of the model to predict each receptor. For comparison we also developed original ResNet-18 without attention module and VGG-19 with and without attention module. The AUC (95% CI), ACC, and F1-score were 0.708 (0.609, 0.808), 0.651, 0.528, respectively, in the HER2 test dataset; 0.785 (0.673, 0.897), 0.845, 0.905, respectively, in the ER test dataset; and 0.706 (0.603, 0.809), 0.678, 0.773, respectively, in the PR test dataset. The proposed model demonstrates superior performance compared to the original ResNet-18 without attention module and VGG-19 with and without attention module. The model has the potential to predict HER2, PR, and especially ER expressions, and thus serve as an adjunctive diagnostic tool for breast cancer.https://doi.org/10.1038/s41598-024-83597-9MammographyDeep learningBreast cancerHistopathology |
| spellingShingle | Shun Zeng Hongyu Chen Rui Jing Wenzhuo Yang Ligong He Tianle Zou Peng Liu Bo Liang Dan Shi Wenhao Wu Qiusheng Lin Zhenyu Ma Jinhui Zha Yonghao Zhong Xianbin Zhang Guangrui Shao Peng Gong An assessment of breast cancer HER2, ER, and PR expressions based on mammography using deep learning with convolutional neural networks Scientific Reports Mammography Deep learning Breast cancer Histopathology |
| title | An assessment of breast cancer HER2, ER, and PR expressions based on mammography using deep learning with convolutional neural networks |
| title_full | An assessment of breast cancer HER2, ER, and PR expressions based on mammography using deep learning with convolutional neural networks |
| title_fullStr | An assessment of breast cancer HER2, ER, and PR expressions based on mammography using deep learning with convolutional neural networks |
| title_full_unstemmed | An assessment of breast cancer HER2, ER, and PR expressions based on mammography using deep learning with convolutional neural networks |
| title_short | An assessment of breast cancer HER2, ER, and PR expressions based on mammography using deep learning with convolutional neural networks |
| title_sort | assessment of breast cancer her2 er and pr expressions based on mammography using deep learning with convolutional neural networks |
| topic | Mammography Deep learning Breast cancer Histopathology |
| url | https://doi.org/10.1038/s41598-024-83597-9 |
| work_keys_str_mv | AT shunzeng anassessmentofbreastcancerher2erandprexpressionsbasedonmammographyusingdeeplearningwithconvolutionalneuralnetworks AT hongyuchen anassessmentofbreastcancerher2erandprexpressionsbasedonmammographyusingdeeplearningwithconvolutionalneuralnetworks AT ruijing anassessmentofbreastcancerher2erandprexpressionsbasedonmammographyusingdeeplearningwithconvolutionalneuralnetworks AT wenzhuoyang anassessmentofbreastcancerher2erandprexpressionsbasedonmammographyusingdeeplearningwithconvolutionalneuralnetworks AT ligonghe anassessmentofbreastcancerher2erandprexpressionsbasedonmammographyusingdeeplearningwithconvolutionalneuralnetworks AT tianlezou anassessmentofbreastcancerher2erandprexpressionsbasedonmammographyusingdeeplearningwithconvolutionalneuralnetworks AT pengliu anassessmentofbreastcancerher2erandprexpressionsbasedonmammographyusingdeeplearningwithconvolutionalneuralnetworks AT boliang anassessmentofbreastcancerher2erandprexpressionsbasedonmammographyusingdeeplearningwithconvolutionalneuralnetworks AT danshi anassessmentofbreastcancerher2erandprexpressionsbasedonmammographyusingdeeplearningwithconvolutionalneuralnetworks AT wenhaowu anassessmentofbreastcancerher2erandprexpressionsbasedonmammographyusingdeeplearningwithconvolutionalneuralnetworks AT qiushenglin anassessmentofbreastcancerher2erandprexpressionsbasedonmammographyusingdeeplearningwithconvolutionalneuralnetworks AT zhenyuma anassessmentofbreastcancerher2erandprexpressionsbasedonmammographyusingdeeplearningwithconvolutionalneuralnetworks AT jinhuizha anassessmentofbreastcancerher2erandprexpressionsbasedonmammographyusingdeeplearningwithconvolutionalneuralnetworks AT yonghaozhong anassessmentofbreastcancerher2erandprexpressionsbasedonmammographyusingdeeplearningwithconvolutionalneuralnetworks AT xianbinzhang anassessmentofbreastcancerher2erandprexpressionsbasedonmammographyusingdeeplearningwithconvolutionalneuralnetworks AT guangruishao anassessmentofbreastcancerher2erandprexpressionsbasedonmammographyusingdeeplearningwithconvolutionalneuralnetworks AT penggong anassessmentofbreastcancerher2erandprexpressionsbasedonmammographyusingdeeplearningwithconvolutionalneuralnetworks AT shunzeng assessmentofbreastcancerher2erandprexpressionsbasedonmammographyusingdeeplearningwithconvolutionalneuralnetworks AT hongyuchen assessmentofbreastcancerher2erandprexpressionsbasedonmammographyusingdeeplearningwithconvolutionalneuralnetworks AT ruijing assessmentofbreastcancerher2erandprexpressionsbasedonmammographyusingdeeplearningwithconvolutionalneuralnetworks AT wenzhuoyang assessmentofbreastcancerher2erandprexpressionsbasedonmammographyusingdeeplearningwithconvolutionalneuralnetworks AT ligonghe assessmentofbreastcancerher2erandprexpressionsbasedonmammographyusingdeeplearningwithconvolutionalneuralnetworks AT tianlezou assessmentofbreastcancerher2erandprexpressionsbasedonmammographyusingdeeplearningwithconvolutionalneuralnetworks AT pengliu assessmentofbreastcancerher2erandprexpressionsbasedonmammographyusingdeeplearningwithconvolutionalneuralnetworks AT boliang assessmentofbreastcancerher2erandprexpressionsbasedonmammographyusingdeeplearningwithconvolutionalneuralnetworks AT danshi assessmentofbreastcancerher2erandprexpressionsbasedonmammographyusingdeeplearningwithconvolutionalneuralnetworks AT wenhaowu assessmentofbreastcancerher2erandprexpressionsbasedonmammographyusingdeeplearningwithconvolutionalneuralnetworks AT qiushenglin assessmentofbreastcancerher2erandprexpressionsbasedonmammographyusingdeeplearningwithconvolutionalneuralnetworks AT zhenyuma assessmentofbreastcancerher2erandprexpressionsbasedonmammographyusingdeeplearningwithconvolutionalneuralnetworks AT jinhuizha assessmentofbreastcancerher2erandprexpressionsbasedonmammographyusingdeeplearningwithconvolutionalneuralnetworks AT yonghaozhong assessmentofbreastcancerher2erandprexpressionsbasedonmammographyusingdeeplearningwithconvolutionalneuralnetworks AT xianbinzhang assessmentofbreastcancerher2erandprexpressionsbasedonmammographyusingdeeplearningwithconvolutionalneuralnetworks AT guangruishao assessmentofbreastcancerher2erandprexpressionsbasedonmammographyusingdeeplearningwithconvolutionalneuralnetworks AT penggong assessmentofbreastcancerher2erandprexpressionsbasedonmammographyusingdeeplearningwithconvolutionalneuralnetworks |