Transformer-Based Radiomics for Predicting Breast Tumor Malignancy Score in Ultrasonography

Breast cancer must be detected early to reduce the mortality rate. Ultrasound images can make it easier for clinician diagnose cases of dense breasts. This study presents a deep vision transformer-based approach predicting breast malignancy scores from ultrasound images. In particular, various state-of-the-art transformers such as BEiT, CaiT, Swin, XCiT, and Vis-Former are adapted trained extract robust radiomics classify tumors in benign or malignant. The best-performing model is used predict score each input image. Experimental results revealed that proposed achieves promising detection malignant on