Related Experiment Video
Updated: Aug 6, 2026
06:03
Integrating Augmented Reality Tools in Breast Cancer Related Lymphedema Prognostication and Diagnosis
Published on: February 6, 2020
Enhancing lymph node metastases assessment in breast cancer post-neoadjuvant therapy using artificial
Yan Ding1, Juan Yu1,2, Min Liu3
1Department of Pathology, The Fourth Hospital of Hebei Medical University, Shijiazhuang, Hebei Province, China.
Breast Cancer Research : BCR
|July 21, 2026
Summary
An AI model accurately classifies breast cancer lymph node metastasis after neoadjuvant therapy (NAT). This artificial intelligence tool improves diagnostic efficiency and reduces pathologist workload for better patient outcomes.
Area of Science:
- Oncology
- Computational Pathology
- Artificial Intelligence in Medicine
Background:
- Neoadjuvant therapy (NAT) is vital for locally advanced breast cancer, but assessing lymph node status post-NAT is challenging due to histological alterations.
- Current methods like immunohistochemistry (IHC) struggle with precise classification of tumor cell presence, including isolated tumor cells (ITCs), micro-metastases, and macro-metastases.
- Accurate post-NAT lymph node assessment is critical for treatment response evaluation and patient management.
Purpose of the Study:
- To develop and validate an artificial intelligence (AI)-driven model for precise classification of lymph node metastasis status in breast cancer patients following NAT.
- To differentiate between negative, ITC, micro-metastasis, and macro-metastasis categories with high accuracy.
- To improve upon existing diagnostic methods for post-NAT lymph node assessment.
Main Methods:
- A weakly supervised Clustering-constrained Attention Multiple Instance Learning (CLAM) framework was employed, analyzing 7764 lymph node whole-slide images (WSIs) from seven cohorts.
- The CLAM model identifies critical subregions within WSIs and generates high-resolution interpretability heatmaps.
- Performance was assessed using binary and multi-class metrics, including external validation and a human-AI comparative analysis on 24 patient samples.
Main Results:
- The AI model demonstrated high performance with an AUROC of 0.97 for binary classification and an overall accuracy of 0.8436 for multi-class differentiation.
- External validation confirmed the model's robustness across diverse datasets.
- In a direct comparison, the AI model outperformed junior pathologists, reducing diagnostic discrepancies by 83%.
Conclusions:
- A robust AI model has been established for accurate and efficient assessment of lymph node metastasis in breast cancer post-NAT.
- The AI model automates the classification process, significantly reducing the workload for pathologists.
- This advancement holds promise for improving diagnostic accuracy and streamlining patient management in breast cancer care.
