Transfer learning drives automatic HER2 scoring on HE-stained WSIs for breast cancer: a multi-cohort study
TL-PA shows robust generalization for HER2 scoring with minimal data; however, it still inadequately capture intra-class variability. This indicates that future deep-learning endeavors should incorporate more detailed annotations to better align the model's focus with the reasoning of pathologists.
Authors
- Xiaoping Li
- Chuqian Lei (ORCID: https://orcid.org/0000-0002-5570-1758)
- Chan Lai
- Chaoran Qiu
- Hui Huang (ORCID: https://orcid.org/0000-0002-3130-3456)
- Yiwen Zhang (ORCID: https://orcid.org/0000-0002-1193-2263)
- Weibin Zhang
- Zhiquan Lin
- Tian Qiu (ORCID: https://orcid.org/0000-0002-8608-5863)
- Weiwen Li (ORCID: https://orcid.org/0000-0001-9533-6575)
- Shaofei Shen (ORCID: https://orcid.org/0000-0002-7161-3838)
Institutions
- Shanxi Agricultural University (CN)
- Shanxi University (CN)
- Shanxi University of Traditional Chinese Medicine (CN)
- Jiangmen Central Hospital (CN)
- Hangzhou Dianzi University (CN)
Publication Details
- Journal
- Breast Cancer Research
- Published
- 2025-04-23
- DOI
- https://doi.org/10.1186/s13058-025-02008-7
- Primary Topic
- AI in cancer detection
- Type
- article
- Field-Weighted Citation Impact
- 0.00
Funders
- National Natural Science Foundation of China
- Hangzhou Dianzi University
- Wuyi University