Preoperative Prediction of Axillary Lymph Node Metastasis in Breast Cancer Based on Intratumoral and Peritumoral DCE-MRI Radiomics Nomogram

Figure 1

Workflows for the necessary steps in the current study. Layer-by-layer manual segmentation of tumors was performed on DCE-MRI images, and the tumor circumference was selected to expand outward 5 mm for semiautomatic profiling, with manual adjustment of the confirmed profiling range. Imaging histology features were extracted from DCE-MRI images of tumor circumference to quantify tumor strength, shape, and texture. In terms of feature selection, the features extracted were selected by using interobserver and intraobserver reliability assessment and LASSO, respectively. The signature of the radiation cohort was constructed from a linear combination of selected features. The performance of the predictive model was assessed by the area under the subject’s working characteristic (ROC) curve. To provide a more comprehensible measurement of results, we developed a nomogram personalized assessment tool that evaluates the fitting excellence of column lines by calibrating curves and analyses the clinical utility of column lines by decision curves.