Early prediction of pathological complete response after neoadjuvant chemotherapy for breast cancer by Delta radiomics based on dynamic contrast-enhanced MRI
10.3969/j.issn.1002-1671.2025.10.015
- VernacularTitle:基于动态对比增强MRI的Delta影像组学对乳腺癌新辅助化疗后病理完全缓解的早期预测
- Author:
Jun HUANG
1
;
Aizhen MA
;
Jing YANG
;
Xiao'an ZHANG
Author Information
1. 河南科技大学临床医学院 河南科技大学第一附属医院影像中心,河南 洛阳 471003;河南科技大学第二附属医院影像中心,河南 洛阳 471000
- Publication Type:Journal Article
- Keywords:
breast cancer;
neoadjuvant chemotherapy;
dynamic contrast-enhanced magnetic resonance imaging;
Delta radiomics;
pathological complete response
- From:
Journal of Practical Radiology
2025;41(10):1663-1668
- CountryChina
- Language:Chinese
-
Abstract:
Objective To explore the value of Delta radiomics based on dynamic contrast-enhanced magnetic resonance imaging(DCE-MRI)in the early prediction of pathological complete response(pCR)after neoadjuvant chemotherapy(NAC)for breast cancer.Methods The MRI and clinical data of 107 patients with breast cancer and received NAC treatment were retrospectively collected.The patients were randomly divided into training set(74 cases)and test set(33 cases)in a ratio of 7︰3.The volume of interest(VOI)were marked,and the radiomics features(Pre-features,Post-features)were extracted from the peak images of DCE-MRI before NAC and after the second cycle of NAC.The Delta features were obtained by subtracting Post-features from Pre-features.Subsequently,six machine learning methods including random forest(RF),multi-layer perceptron(MLP),extreme gradient boosting(XGBoost),light gradient boosting machine(LightGBM),K-nearest neighbor(KNN)and logistic regression(LR)were used to consult the model,respectively.The receiver operating characteristic(ROC)curve was used to evaluate the model performance,and the machine learning method and radiomics feature set with the highest area under the curve(AUC)in the test set were selected for further analysis.Univariate and multivariate logistic regression analyses were performed to identify independent clinical predictors and the machine learning method with the best radiomics model was used to construct the clinical model.Combined with tumor radiomics features and independent clinical predictors,the best machine learning method was used to construct a combined model,and the performance of the model was evaluated.Results Progesterone receptor(PR)status was identified as an independent predictor of NAC efficacy in breast cancer.The model established using the RF machine learning method based on the Delta features had the highest AUC of 0.926 in the test set.The AUC of the combined model in the test set was 0.957,which was higher than that of the Pre-radiomics model,Post-radiomics model,Delta-radiomics model,and clinical-features model.Conclusion The combined model of Delta features and clinical features has a good performance in predicting pCR after NAC for breast cancer.
