Noninvasive Preoperative Assessment of Sentinel Lymph Node Metastasis in Breast Cancer Using Super-Resolution
Ruifang Guo1, Xi Liu1, Junfeng Zhao1
1Department of Ultrasound, Beijing Friendship Hospital, Capital Medical University, Beijing, China.
Journal of Clinical Ultrasound : JCU
|June 29, 2026
Summary
Super-resolution ultrasound microvascular imaging (SR-UMI) shows promise for predicting breast cancer sentinel lymph node involvement. SR-UMI significantly improved prediction accuracy compared to clinical factors alone.
Area of Science:
- Medical Imaging
- Oncology
- Vascular Biology
Background:
- Sentinel lymph node (SLN) biopsy is standard for staging breast cancer.
- Accurate preoperative prediction of SLN involvement can optimize surgical planning.
- Noninvasive imaging methods are needed to improve diagnostic accuracy.
Purpose of the Study:
- To develop and validate a super-resolution ultrasound microvascular imaging (SR-UMI) model for predicting breast cancer SLN involvement.
- To assess the incremental value of SR-UMI over traditional clinical variables.
- To evaluate the model's performance in a prospective cohort.
Main Methods:
- Prospective study of 162 breast cancer patients undergoing contrast-enhanced SR-UMI.
- SR-UMI data analyzed for microvascular features (98 features).
- Elastic net-regularized logistic regression models compared clinical-only, SR-UMI-only, and combined predictors using cross-validation.
Main Results:
- SR-UMI-only model achieved an AUC of 0.800, significantly outperforming the clinical-only model (AUC 0.662).
- The combined model (SR-UMI + clinical) showed similar performance to SR-UMI-only (AUC 0.802).
- At the optimal threshold, the combined model achieved 78.2% sensitivity and 76.6% specificity.
Conclusions:
- SR-UMI microvascular features significantly enhance the prediction of breast cancer SLN involvement.
- The SR-UMI model offers substantial improvement over clinical variables alone.
- External validation is recommended for broader applicability.
