Article details

Research area
Natural language & AI

Society for Imaging Informatics in Medicine conference 2019 (SIIM’19)


Jen-Tang Lu, Rupert Brooks, Varun Buch, Ryan King, Stefan Hahn, Jin Chen, Gopal Kotecha, Bernardo C. Bizzo, Katherine P. Andriole, Joel Pinto, Brian Ghoshhajra, Mark H. Michalski

DeepAAA: Automated Detection of Abdominal Aortic Aneurysms using Deep Learning


Untreated abdominal aortic aneurysms (AAAs) tend to grow and eventually may rupture with mortality rates exceeding 90%. As most of AAAs are asymptomatic until onset of bleeding, incidental finding of AAAs becomes critical. However, on routine abdominal computed tomography (CT) exams, only 65% of AAAs are incidentally identified. This low reporting rate makes it difficult to provide patients optimal treatment. To address the issue, this study aims to develop a deep learning-based system (DeepAAA) for automated AAA detection on contrast and non-contrast CT scans.

Read/download now