CONSTRUCTION OF A BAYESIAN NETWORK FOR MAMMOGRAPHIC DIAGNOSIS OF BREAST-CANCER

Bayesian networks use the techniques of probability theory to reason u nder uncertainty, and have become an important formalism for medical d ecision support systems. We describe the development and validation of a Bayesian network (MammoNet) to assist in mammographic diagnosis of breast cancer. MammoNet integrates five patient-history features, two physical findings, and 15 mammographic features extracted by experienc ed radiologists to determine the probability of malignancy. We outline the methods and issues in the system's design, implementation, and ev aluation. Bayesian networks provide a potentially useful tool for mamm ographic decision support. (C) 1997 Elsevier Science Ltd.