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

To establish a radiomics nomogram model integrating features extracted from ultrafast dynamic contrast-enhanced magnetic resonance imaging (ultrafast DCE-MRI) and apparent diffusion coefficient (ADC) quantitative maps, with clinicopathological and imaging risk characteristics, for preoperative predicting the status of axillary lymph node (ALN) in clinical T1-2 stage invasive breast cancers. A total of 201 patients were retrospectively enrolled and randomly divided into training and testing cohorts in a 7:3 ratio. Radiomics features were extracted from quantitative maps (Ktrans, Kep, Ve, and ADC). Model based on combinations of sequence was constructed by support vector machine (SVM) classifier, and the radiomics signature was calculated. Univariate and multivariate logistic regression analyses were used to determine the independent risk factors for constructing the nomogram model. The diagnostic performance of the model was assessed by receiver operating characteristic (ROC) curve analysis and calibration curve. The radiomics signature was constructed using 11 selected radiomics features, achieving areas under the ROC curve (AUCs) of 0.831 and 0.821 in the training and testing cohort, respectively. The radiomics nomogram model incorporating the radiomics signature, multifocality, and MRI report ALN status exhibited superior performance with AUC values of 0.890 and 0.905 in the training and testing cohort, respectively. The calibration curve results indicated that the nomogram predictions displayed promising clinical utility. The ultrafast DCE-MRI and ADC quantitative maps based radiomics and the nomogram showed valuable predictive capabilities for evaluating ALN status in clinical T1-2 stage invasive breast cancers.