MRI-based radiomics nomogram predicts axillary lymph node status before breast cancer surgery

Researchers have developed a radiomics nomogram that combines quantitative features from ultrafast dynamic contrast-enhanced MRI (DCE-MRI) and apparent diffusion coefficient (ADC) maps with clinicopathological and imaging risk characteristics to predict axillary lymph node (ALN) status before surgery in patients with clinical T1-2 stage invasive breast cancer.

The study retrospectively enrolled 201 patients, who were randomly divided into training and testing cohorts in a 7:3 ratio. Radiomics features were extracted from quantitative maps including Ktrans, Kep, Ve, and ADC. Models based on different combinations of sequences were built using a support vector machine (SVM) classifier, from which a radiomics signature was calculated. Univariate and multivariate logistic regression analyses were then used to identify independent risk factors for inclusion in the final nomogram model.

The radiomics signature alone was constructed from 11 selected features and achieved areas under the ROC curve (AUCs) of 0.831 in the training cohort and 0.821 in the testing cohort. The full nomogram, which incorporated the radiomics signature together with multifocality and the MRI report of ALN status, performed better still, with AUC values of 0.890 in the training cohort and 0.905 in the testing cohort. Calibration curve results indicated that the nomogram’s predictions showed promising clinical utility, according to the authors.

Combining these quantitative imaging features with clinical variables such as multifocality is the defining design feature of the nomogram, which is intended for preoperative rather than intraoperative assessment.

The authors conclude that the radiomics signature derived from ultrafast DCE-MRI and ADC quantitative maps, and the nomogram built upon it, showed valuable predictive capability for evaluating ALN status in clinical T1-2 stage invasive breast cancers.

Subject of Research: Preoperative prediction of axillary lymph node status in clinical T1-2 invasive breast cancer using a radiomics nomogram based on ultrafast DCE-MRI and ADC maps

Article Title: Radiomics nomogram based on ultrafast DCE-MRI and ADC map: preoperative evaluation of axillary lymph node status in clinical T1-2 invasive breast cancer

Article References: Zeng, F., Tang, X., Lin, L., Chen, C., Wu, Y., Wang, Z., Chen, M., Chen, L., Chen, L., Wu, P., Wang, C., & Xue, Y. (2026). Radiomics nomogram based on ultrafast DCE-MRI and ADC map: preoperative evaluation of axillary lymph node status in clinical T1-2 invasive breast cancer. BMC Cancer. https://doi.org/10.1186/s12885-026-16845-y

Image Credits: AI Generated

DOI: 10.1186/s12885-026-16845-y

Keywords: breast cancer, radiomics, nomogram, ultrafast DCE-MRI, ADC map, axillary lymph nodes, machine learning, SVM, diagnostic prediction, MRI, oncology, medical imaging