Enhancing lymph node metastases assessment in breast cancer post-neoadjuvant therapy using artificial intelligence-driven diagnostics
8 affiliations
- 1Department of Pathology, The Fourth Hospital of Hebei Medical University
- 2Medical Affairs Department, Betrue AI Lab
- 3Department of Pathology, Affiliated Hospital of Hebei University
- 4Department of Pathology, Xingtai People’s Hospital
- 5Department of Pathology, First Hospital of Qinhuangdao
- 6Department of Pathology, Handan First Hospital
- 7Department of Pathology, Cangzhou Integrated Traditional Chinese and Western Medicine Hospital
- 8Department of Pathology, The Ohio State University Wexner Medical Center
Abstract
Background Neoadjuvant therapy (NAT) is crucial for locally advanced breast cancer, but post-NAT lymph node assessment is challenging due to histological changes. Current methods like immunohistochemistry (IHC) are labor-intensive and imprecise in distinguishing isolated tumor cells (ITCs), micro-metastases (Micro), and macro-metastases (Macro). We aimed to develop and validate an AI-driven model for precise classification of lymph node metastasis status (negative, ITC, Micro, Macro) in breast cancer patients post-NAT. Methods We used a weakly supervised Clustering-constrained Attention Multiple Instance Learning (CLAM) framework to analyze 7764 lymph node samples from 7 cohorts. The CLAM model identifies high-diagnostic-value subregions within whole-slide images (WSIs) and generates high-resolution interpretability heatmaps. Performance was evaluated using binary and multi-class metrics, with external validation on diverse datasets. A human-AI comparative analysis was conducted on 24 patient-derived lymph node sections. Results The AI model achieved an AUROC of 0.97 (95% CI: 0.962–0.977) in binary classification and an overall accuracy of 0.8436 (95% CI: 0.8282–0.8562) for multi-class differentiation. In the human-AI comparison, the model outperformed junior pathologists, reducing diagnostic discrepancies by 83%. Conclusion This study establishes a robust AI model that significantly improves the accuracy and efficiency of post-NAT lymph node metastasis assessment in breast cancer, automating classification and reducing pathologist workload.
Related
- Artificial Intelligence–Driven Multimodal Imaging for Cancer During Pregnancy: Advances in Maternal–Fetal Diagnostics and Precision Oncology
- Artificial Intelligence in Rapid Diagnostics for Antimicrobial Resistance: From Bench to Bedside
- An assisted telemedicine model with digital diagnostics and artificial intelligence-supported clinical decision in hard-to-reach rural communities: a community-based mixed-methods study from Rajasthan, India
- Artificial intelligence assisted genomic profiling of ovarian cancer liver metastases
- Theranostic Imaging Agents: Integrating Molecular Diagnostics, Precision Therapeutics, and Artificial Intelligence
BibTeX
@article{ding2026enhancing,
title = {{Enhancing lymph node metastases assessment in breast cancer post-neoadjuvant therapy using artificial intelligence-driven diagnostics}},
author = {Yan Ding and Juan Yu and Min Liu and Xiangyu Liu and Ling Kang and Liujing Huang and Jinze Li and Yanan Wang and Xin Xu and Min Zhao and Ping Wei and Shuangbiao Li and Zaibo Li and Yueping Liu},
journal = {Breast Cancer Research},
year = {2026},
month = jul,
doi = {10.1186/s13058-026-02349-x},
url = {https://doi.org/10.1186/s13058-026-02349-x},
note = {CC BY},
}
RIS
TY - JOUR
TI - Enhancing lymph node metastases assessment in breast cancer post-neoadjuvant therapy using artificial intelligence-driven diagnostics
AU - Yan Ding
AU - Juan Yu
AU - Min Liu
AU - Xiangyu Liu
AU - Ling Kang
AU - Liujing Huang
AU - Jinze Li
AU - Yanan Wang
AU - Xin Xu
AU - Min Zhao
AU - Ping Wei
AU - Shuangbiao Li
AU - Zaibo Li
AU - Yueping Liu
PY - 2026
DA - 2026/07/01
JO - Breast Cancer Research
DO - 10.1186/s13058-026-02349-x
UR - https://doi.org/10.1186/s13058-026-02349-x
AB - Background Neoadjuvant therapy (NAT) is crucial for locally advanced breast cancer, but post-NAT lymph node assessment is challenging due to histological changes. Current methods like immunohistochemistry (IHC) are labor-intensive and imprecise in distinguishing isolated tumor cells (ITCs), micro-metastases (Micro), and macro-metastases (Macro). We aimed to develop and validate an AI-driven model for precise classification of lymph node metastasis status (negative, ITC, Micro, Macro) in breast cancer patients post-NAT. Methods We used a weakly supervised Clustering-constrained Attention Multiple Instance Learning (CLAM) framework to analyze 7764 lymph node samples from 7 cohorts. The CLAM model identifies high-diagnostic-value subregions within whole-slide images (WSIs) and generates high-resolution interpretability heatmaps. Performance was evaluated using binary and multi-class metrics, with external validation on diverse datasets. A human-AI comparative analysis was conducted on 24 patient-derived lymph node sections. Results The AI model achieved an AUROC of 0.97 (95% CI: 0.962–0.977) in binary classification and an overall accuracy of 0.8436 (95% CI: 0.8282–0.8562) for multi-class differentiation. In the human-AI comparison, the model outperformed junior pathologists, reducing diagnostic discrepancies by 83%. Conclusion This study establishes a robust AI model that significantly improves the accuracy and efficiency of post-NAT lymph node metastasis assessment in breast cancer, automating classification and reducing pathologist workload.
SN - 1465-542X
SN - 1465-5411
ER -