The value of different machine learning radiomics based on DCE-MRI in predicting axillary lymph node status of breast cancer - Research paper summary

Sep 30, 2026

Uploaded by Jacob

Accurate preoperative assessment of axillary lymph node status is essential for treatment selection in primary breast cancer, yet current approaches are either invasive or insufficiently precise. This study evaluates machine learning models that combine dynamic contrast-enhanced MRI (DCE-MRI) radiomics with clinicopathologic variables to predict different AJCC lymph node stages (N0, N1, N2–3). Using retrospective data from 605 patients, models were trained and tested with feature selection, logistic regression, ROC evaluation, and decision curve analysis to confirm improved diagnostic accuracy and clinical utility for axillary lymph node metastasis.

LanguageEnglish

CategoryHealthcare

FormatPDF

Pages14

Reading Time35 min

AccessFree to download

Background

Methods

Results

Conclusions

Highlights

Key findings

What is known and what is new?

What is the implication, and what should change now?

The value of different machine learning radiomics based on DCE MRI in predicting axillary lymph node status of breast cancer Han Wang^, Li Gong Department of Ultrasound Medicine, Nanjing Drum Tower Hospital, Affiliated Hospital of Nanjing University Medical School, Nanjing, China Contributions: (I) Conception and design: H Wang; (II) Administrative support: L Gong; (III) Provision of study materials or patients: H Wang; (IV) Collection and assembly of data: Both authors; (V) Data analysis and interpretation: Both authors; (VI) Manuscript writing: Both authors; (VII) Final approval of manuscript: Both authors. Correspondence to: Li Gong, MMed. Department of Ultrasound Medicine, Nanjing Drum Tower Hospital, Affiliated Hospital of Nanjing University Medical School, 321 Zhongshan Street, Nanjing 21009, China. Email: [gongli2010@163.com](gongli2010@163.com). Background: The accurate preoperative assessment of axillary lymph node (ALN) status is critical for therapeutic decision-making in primary breast cancer (BC), yet current methods are either invasive or lack precision. The objective of this study was to investigate the performance of machine learning models based on dynamic contrast-enhanced magnetic resonance imaging (MRI), in conjunction with clinicopathologic data, in predicting different American Joint Committee on Cancer (AJCC) lymph node (N) stages in patients with BC. Methods: The data of 605 BC patients were retrospectively analyzed and separated into training and test sets. Following dimensionality reduction and feature selection, a predictive model was established via machine learning techniques. Clinicopathologic features were assessed through both univariable and multivariable logistic regressions (LRs) to select variables for constructing clinical models. The optimal radiomics and clinical models were identified via receiver operating characteristic (ROC) curve analysis and integrated into a combined model. The clinical utility of this combined model was evaluated via decision curve analysis (DCA), which confirmed its superior diagnostic accuracy in detecting axillary lymph node metastasis (ALNM) . Results: The combined model yielded area under the curve (AUC) values of 0.890 and 0.854 in the training and test sets, respectively. Additionally, in differentiating the N1 group from the N2–3 group, the combined model showed strong performance, with AUC values of 0.973 and 0.835 in the training and test sets, respectively. Moreover, the model effectively classified the N0, N1, and N2–3 groups, achieving a micro-AUC of 0.861 and a macro-AUC of 0.812. Conclusions: The integration of radiomics features with clinicopathologic characteristics provides a robust predictive tool for ALNM, potentially offering a noninvasive and effective approach for clinical decision making. Keywords: Machine learning; radiomics; breast cancer (BC); axillary lymph node (ALN); clinicopathology Submitted Jul 02, 2025. Accepted for publication Oct 13, 2025. Published online Dec 29, 2025. doi: 10.21037/tcr-2025-1418 View this article at: [https://dx.doi.org/10.21037/tcr-2025-1418](https://dx.doi.org/10.21037/tcr-2025-1418) ^ ORCID: 0009-0009-7696-992X. © AME Publishing Company. Transl Cancer Res 2025;14(12):8475-8488 | [https://dx.doi.org/10.21037/tcr-2025-1418](https://dx.doi.org/10.21037/tcr-2025-1418) 8476 Wang and Gong. Radiomics based on DCE-MRI predict ALN status of BC Introduction Breast cancer (BC) is the leading cause of cancer-related mortality among women globally (1) . Importantly, the prognosis of the disease vary significantly across differentlymph node stages (N stages), necessitating different surgical and adjuvant treatment approaches (2,3) . Initially, patients with N-stage disease are evaluated through either sentinel lymph node biopsy (SLNB) or axillary lymph node dissection (ALND) on the basis of their individual circumstances (4,5) . The National Comprehensive Cancer Network (NCCN)

FAQs

What problem does the study address in breast cancer care?

It addresses the need for accurate preoperative prediction of axillary lymph node status, which is critical for selecting appropriate therapy but often limited by invasive or imprecise methods.

How are radiomics features and clinical data used in the predictive models?

The study builds machine learning models using DCE-MRI radiomics features after dimensionality reduction and feature selection, and integrates selected clinicopathologic variables via univariable and multivariable logistic regression.

How well did the combined model perform for different lymph node stages?

The combined model achieved AUCs of 0.890 (training) and 0.854 (test) and showed strong performance for distinguishing N1 from N2–3 with AUCs of 0.973 (training) and 0.835 (test), alongside multiclass discrimination (micro-AUC 0.861; macro-AUC 0.812).

What is the clinical significance of the model based on decision curve analysis?

Decision curve analysis indicated superior clinical utility compared with radiomics-only or clinical-only approaches, supporting its potential to guide treatment decisions and reduce unnecessary invasive biopsies.

Sep 30, 2026

Uploaded by Jacob