Machine learning prediction of HER2-low expression in breast cancers based on hematoxylin–eosin-stained slides

• Published in: Breast cancer research : BCR Vol. 27; no. 1; pp. 57 - 10
• Main Authors: Du, Jun, Shi, Jun, Sun, Dongdong, Wang, Yifei, Liu, Guanfeng, Chen, Jingru, Wang, Wei, Zhou, Wenchao, Zheng, Yushan, Wu, Haibo
• Format: Journal Article
• Language: English
• Published: England BioMed Central Ltd 18.04.2025; BioMed Central; BMC
• Subjects: Artificial intelligence; Biomarkers, Tumor; Breast cancer; Breast Neoplasms - diagnosis; Breast Neoplasms - genetics; Breast Neoplasms - metabolism; Breast Neoplasms - pathology; Cancer; Care and treatment; Clinical medicine; Clinical trials; Datasets; Deep learning; Developing countries; Eosine Yellowish-(YS) - chemistry; ErbB-2 protein; Female; Gene amplification; Genetic aspects; Graph representations; Hematoxylin - chemistry; Hematoxylin–eosin; HER2; Humans; Hybridization; Immunohistochemistry; Immunohistochemistry - methods; LDCs; Learning algorithms; Machine Learning; Neural networks; Pathology; Patients; Pertuzumab; Prediction; Receptor, ErbB-2 - genetics; Receptor, ErbB-2 - metabolism; ROC Curve; Semantics; Staining and Labeling - methods; Tumors; British Columbia; China; Breast cancer; HER2; Hematoxylin–eosin; Machine learning; Prediction

Treatment with HER2-targeted therapies is recommended for HER2-positive breast cancer patients with HER2 gene amplification or protein overexpression. Interestingly, recent clinical trials of novel HER2-targeted therapies demonstrated promising efficacy in HER2-low breast cancers, raising the prospe...