Computer programs that automatically recognize anatomy and pathology in Magnetic Resonance Imaging (MRI) and Computed Tomography (CT) scans have shown some preliminary success. Existing systems recognize, with varying accuracy, a small number of anatomical features and, again with varying degrees of accuracy, identify a small number of pathologies. The programs draw on methods from pattern recognition, artificial intelligence, and computer vision. In this article we review existing systems for automated recognition of brain anatomy and pathology in MRI and CT images, the methods used, their performance, and directions for future research.
