Prospective Study of MRI-based Radiomics in Breast Cancer

The written informed consents for the participation into this prospective study is done in participants who are diagnosed with invasive breast cancer before surgery with curative intent. After enrollment, breast MRI is done for particpants and diffusion-, T2-, and dynamic contrast-enhanced T1-weighted images are extracted from the results of MRI of participants. The tumor tissue samples are archived from participants before surgery through core-needle biopsy and the immunohistochemistry is done to assess the components of tumor microenvironment including intraturmoal and peritumoral components through volume histochemical assessment. Radiomic feature selection and prediction model building are performed using the least absolute shrinkage and selection operator and linear regression analysis, respectively. The investigators apply fivefold cross-validation to prevent overfitting, and the performance is evaluated using the area under the receiver operating characteristic curve (AUC) and the DeLong test.