BI-RADS-NET-V2: A Composite Multi-Task Neural Network For Computer-Aided Diagnosis Of Breast Cancer In Ultrasound Images With Semantic And Quantitative Explanations

Computer-aided Diagnosis (CADx) based on explainable artificial intelligence (XAI) can gain the trust of radiologists and effectively improve diagnosis accuracy consultation efficiency. This paper proposes BI-RADS-Net-V2, a novel machine learning approach for fully automatic breast cancer in ultrasound images. The BI-RADS-Net-V2 accurately distinguish malignant tumors from benign ones provides both semantic quantitative explanations. explanations are provided terms clinically proven morphological features used by clinicians reporting mass findings, i.e., Breast Imaging Reporting Data System (BI-RADS). experiments 1,192 Ultrasound (BUS) images indicate that proposed method improves taking full advantage medical knowledge BI-RADS while providing decision.