Value of predicting HER-2 and Ki-67 expression status in breast cancer based on multiparametric MRI intratumor combined with peritumor radiomics

Motivation: There is an urgent need to find a non-invasive method that can accurately predict HER-2 and Ki-67 expression status in breast cancer. Goal(s): Establishment of multiparametric MRI intratumor combined with peritumor radiomics models for preoperative prediction of HER-2 and Ki-67 expression status in breast cancer. Approach: A two-center retrospective study. Results: A random forest （RF） machine learning algorithm was used to construct eight radiomics models for preoperative predict HER-2 and Ki-67 expression status in breast cancer: intratumoral radiomics models, intratumoral combined with peritumoral (3-mm) radiomics models, and multisequence fusion radiomics models. Impact: Accurate preoperative prediction of HER-2 and Ki-67 expression status in breast cancer is expected to provide a reference for precise and personalized treatment decisions in later stages of clinical practice.