Related Experiment Video
Updated: Jun 28, 2026
15:48
Tracking the Mammary Architectural Features and Detecting Breast Cancer with Magnetic Resonance Diffusion Tensor Imaging
Published on: December 15, 2014
22.3K
Deep learning-based breast MRI for predicting axillary lymph node metastasis: a systematic review and meta-analysis
Chia-Fen Lee1,2, Joseph Lin3,4,5, Yu-Len Huang6
1Department of Radiology, Changhua Christian Hospital, Changhua, Taiwan.
Summary
Deep learning algorithms show promise in predicting breast cancer axillary lymph node metastases using MRI scans. This AI approach can improve diagnostic accuracy and potentially reduce the need for invasive surgical procedures.
Area of Science:
- Medical Imaging
- Artificial Intelligence
- Oncology
Background:
- Breast cancer diagnosis and staging often involve assessing axillary lymph node status.
- Accurate prediction of lymph node metastases is crucial for treatment planning and patient outcomes.
- Current methods may have limitations, necessitating advanced diagnostic tools.
Purpose of the Study:
- To systematically review and meta-analyze the diagnostic performance of deep learning (DL) algorithms.
- To evaluate DL's accuracy in predicting axillary lymph node metastases from breast MRI.
- To assess the potential of DL in breast cancer management.
Main Methods:
- Systematic literature search of PubMed, MEDLINE, and Embase databases (Jan 2004 - Feb 2025).
- Inclusion of studies using DL on breast MRI for axillary lymph node metastasis prediction.
- Quality assessment using QUADAS-AI and ACCLAIM; pooled diagnostic accuracy calculated via meta-analysis and SROC curves.
Main Results:
- Ten studies were included in the meta-analysis.
- Pooled sensitivity was 0.76 (95% CI, 0.67-0.83) and pooled specificity was 0.81 (95% CI, 0.74-0.87).
- Moderate heterogeneity was observed; the weighted area under the receiver operating characteristic curve (AUC) was 0.788.
Conclusions:
- Deep learning algorithms applied to breast MRI demonstrate promising diagnostic performance for predicting axillary lymph node metastases.
- Integration of DL into clinical practice could enhance decision-making for breast cancer patients.
- This non-invasive method may improve the accuracy of predicting lymph node involvement, potentially reducing unnecessary surgeries.
