Supervised contrastive pre-training models for mammography screening
Abstract Breast cancer is now the most deadly cancer worldwide. Mammography screening is the most effective method for early detection and diagnosis of breast cancer. Due to the lack of labeled mammograms, building an AI system for mammography screening often relies heavily on human-designed data au...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
SpringerOpen
2025-02-01
|
| Series: | Journal of Big Data |
| Subjects: | |
| Online Access: | https://doi.org/10.1186/s40537-025-01075-z |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1823862018187198464 |
|---|---|
| author | Zhenjie Cao Zhuo Deng Zhicheng Yang Jie Ma Lan Ma |
| author_facet | Zhenjie Cao Zhuo Deng Zhicheng Yang Jie Ma Lan Ma |
| author_sort | Zhenjie Cao |
| collection | DOAJ |
| description | Abstract Breast cancer is now the most deadly cancer worldwide. Mammography screening is the most effective method for early detection and diagnosis of breast cancer. Due to the lack of labeled mammograms, building an AI system for mammography screening often relies heavily on human-designed data augmentation, which doesn’t always perform robustly when applied to clinical scenarios. This paper presents a novel framework of Supervised Contrastive Pre-training followed by Supervised Fine-tuning (SCP+SF) for mammography screening. Unlike the previous approaches, the proposed supervised contrastive pre-training does not need a data augmentation module. We apply the SCP+SF framework to two challenging and important mammography screening tasks for breast cancer: mammographic abnormality screening and mammographic malignancy screening. Our extensive experiments on a large-scale dataset show that the supervised contrastive pre-training (SCP) can substantially improve the final model performance compared with the traditional direct supervised training approach. Superior results of AUC and specificity/sensitivity have been achieved on two clinically significant mammographic screening tasks in comparison with previously reported State-Of-The-Art approaches. We believe this work is the first to show that supervised contrastive pre-training (SCP) followed by supervised fine-tuning (SF) can outperform the supervised counterpart on these two critical medical imaging tasks. |
| format | Article |
| id | doaj-art-fdf17e6b9dd04df4ba3ddc92c6aa286b |
| institution | Kabale University |
| issn | 2196-1115 |
| language | English |
| publishDate | 2025-02-01 |
| publisher | SpringerOpen |
| record_format | Article |
| series | Journal of Big Data |
| spelling | doaj-art-fdf17e6b9dd04df4ba3ddc92c6aa286b2025-02-09T12:41:19ZengSpringerOpenJournal of Big Data2196-11152025-02-0112111710.1186/s40537-025-01075-zSupervised contrastive pre-training models for mammography screeningZhenjie Cao0Zhuo Deng1Zhicheng Yang2Jie Ma3Lan Ma4Shenzhen International Graduate School, Tsinghua UniversityShenzhen International Graduate School, Tsinghua UniversityPAII Inc.Shenzhen People’s HospitalShenzhen International Graduate School, Tsinghua UniversityAbstract Breast cancer is now the most deadly cancer worldwide. Mammography screening is the most effective method for early detection and diagnosis of breast cancer. Due to the lack of labeled mammograms, building an AI system for mammography screening often relies heavily on human-designed data augmentation, which doesn’t always perform robustly when applied to clinical scenarios. This paper presents a novel framework of Supervised Contrastive Pre-training followed by Supervised Fine-tuning (SCP+SF) for mammography screening. Unlike the previous approaches, the proposed supervised contrastive pre-training does not need a data augmentation module. We apply the SCP+SF framework to two challenging and important mammography screening tasks for breast cancer: mammographic abnormality screening and mammographic malignancy screening. Our extensive experiments on a large-scale dataset show that the supervised contrastive pre-training (SCP) can substantially improve the final model performance compared with the traditional direct supervised training approach. Superior results of AUC and specificity/sensitivity have been achieved on two clinically significant mammographic screening tasks in comparison with previously reported State-Of-The-Art approaches. We believe this work is the first to show that supervised contrastive pre-training (SCP) followed by supervised fine-tuning (SF) can outperform the supervised counterpart on these two critical medical imaging tasks.https://doi.org/10.1186/s40537-025-01075-zBreast cancerMammography screeningSupervised contrastive pre-training |
| spellingShingle | Zhenjie Cao Zhuo Deng Zhicheng Yang Jie Ma Lan Ma Supervised contrastive pre-training models for mammography screening Journal of Big Data Breast cancer Mammography screening Supervised contrastive pre-training |
| title | Supervised contrastive pre-training models for mammography screening |
| title_full | Supervised contrastive pre-training models for mammography screening |
| title_fullStr | Supervised contrastive pre-training models for mammography screening |
| title_full_unstemmed | Supervised contrastive pre-training models for mammography screening |
| title_short | Supervised contrastive pre-training models for mammography screening |
| title_sort | supervised contrastive pre training models for mammography screening |
| topic | Breast cancer Mammography screening Supervised contrastive pre-training |
| url | https://doi.org/10.1186/s40537-025-01075-z |
| work_keys_str_mv | AT zhenjiecao supervisedcontrastivepretrainingmodelsformammographyscreening AT zhuodeng supervisedcontrastivepretrainingmodelsformammographyscreening AT zhichengyang supervisedcontrastivepretrainingmodelsformammographyscreening AT jiema supervisedcontrastivepretrainingmodelsformammographyscreening AT lanma supervisedcontrastivepretrainingmodelsformammographyscreening |