Truace
Subscribe

A deep learning radiomics model based on multiparametric MRI for predicting the efficacy of neoadjuvant chemotherapy in axillary lymph nodes of breast cancer

Objectives In this study, we developed and validated a deep learning radiomics clinical (DLRC) model integrating multiparametric MRI (mpMRI) and clinical data to predict axillary pathological complete response (apCR) in cN+ breast cancer patients receiving neoadjuvant chemotherapy (NAC). Materials and methods We retrospectively collected data from 221 eligible breast cancer patients. The DLRC model integrated radiomics features and deep learning scores (DL-Scores) derived from mpMRI, together with clinical risk…

By the Truvace desk2 min read1 sourcePeer-reviewedTRV-2026-2229· cite & verify
A deep learning radiomics model based on multiparametric MRI for predicting the efficacy of neoadjuvant chemotherapy in axillary lymph nodes of breast cancer
BREAST CANCER RESEARCH AND DEVELOPMENT by Committee on Appropriations. Public domain

In brief

Researchers built a model that merges radiomics features and deep learning scores from multiparametric breast MRI with clinical risk factors. It was designed to estimate whether cancer in the axillary lymph nodes would be completely cleared after neoadjuvant chemotherapy in patients with node-positive disease.

If the approach holds up, it could give clinicians an auxiliary tool for judging which patients might safely receive less extensive axillary treatment, supporting more personalized care. The evidence comes from a retrospective cohort of 221 patients, so how well it performs in routine practice remains untested.

Main points

  1. The study included 221 eligible breast cancer patients whose data were collected retrospectively.
  2. The model combined radiomics features, deep learning scores from multiparametric MRI, and clinical risk factors.
  3. AUC was 0.949 in the training set and 0.932 in the validation set.
  4. Progesterone receptor, HER2 status, clinical T stage, and time-intensity curve were significantly associated with apCR.
  5. SHAP plots were used to show how the model reached its predictions.

The gain

A deep learning radiomics model combining multiparametric MRI and clinical data predicted axillary pathological complete response after neoadjuvant chemotherapy, potentially helping clinicians identify candidates for axillary de-escalation.

The rundown

The target outcome was axillary pathological complete response in patients with clinically node-positive breast cancer who received neoadjuvant chemotherapy. The authors tested seven classifiers and selected the best-performing one to build the final model.

Performance was assessed with ROC curves, AUC, calibration curves, decision curve analysis, and a confusion matrix. Multivariate analysis linked progesterone receptor, HER2, clinical T stage (OR = 0.258), and time-intensity curve to the outcome.

The authors present the model as support for personalized precision treatment and for identifying candidates for axillary de-escalation. They report following the TRIPOD-AI reporting checklist.

Sources

  1. Peer-reviewedMagnetic Resonance Imaging2026-10-21

The debate