Abstract: Breast cancer is the most common malignant tumor in Chinese women. Early diagnosis, treatment, and prognosis assessment are important clinical problems to be solved. Radiomics is a non-invasive method for high-throughput extraction and analysis of lesion features of images to provide more potential information of tumors, and guide precise diagnosis and treatment. Recently, it has been widely concerned and studied in breast cancer, covering every stage in the care of patients with breast cancer. In terms of diagnosis, the research has reached the level of maturity and gradually proceeded to the clinical setting. Regarding efficacy evaluation and prognostic prediction, although in its infancy, radiomics shows promising potential. This paper mainly reviews the application of radiomics in the diagnosis, response evaluation, and prognosis prediction of breast cancer.