Computer-aided diagnosis in breast MRI: Do adjunct features derived from T2-weighted images improve classification of breast masses?

Abstract

In the field of dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) of breast cancer, current research efforts in computer-aided diagnosis (CADx) are mainly focused on the temporal series of T1-weighted images acquired during uptake of a contrast agent, processing morphological and kinetic information. Although static T2-weighted images are usually part of DCE-MRI protocols, they are seldom used in CADx systems. The aim of this work was to evaluate to what extent T2-weighted images provide complementary information to a CADx system, improving its performance for the task of discriminating benign breast masses from life-threatening carcinomas. In a preliminary study considering 64 masses, inclusion of lesion features derived from T2-weighted images increased the classification performance from Az=0.94 to Az=0.99.