Open Collections

A deep learning framework for wall motion abnormality detection in echocardiograms

University of British Columbia

Coronary Artery Disease (CAD) is the leading cause of morbidity and mortality in developed nations. In patients with acute or chronic obstructive CAD, Echocardiography (ECHO) is the standard-of-care for visualizing abnormal ventricular wall thickening or motion which would be reported as Regional WallMotion Abnormality (RWMA). The accurate identification of regional wall motion abnormalities is essential for cardiovascular assessment and myocardial ischemia, coronary artery disease and myocardial infarction diagnosis. Given the variability and challenges of scoring regional wall motion abnormalities, we propose the development of a platform that can quickly and accurately identify regional and global wall motion abnormalities on echo images. This thesis describes a deep learning-based framework that can aid physicians to utilize ultrasound for wall motion abnormality detection. The framework jointly combines image data and patient diagnostic information to determine both global and clinically-standard 16 regional wall motion labels. We validate the approach on a large cohort of echo studies obtained from 953 patients. We then report the performance of the proposed framework in the detection of wall motion abnormality. An average accuracy of 69.2% for the 16 regions and an average accuracy of 69.5% for global wall motion abnormality were achieved. To the best of our knowledge, our proposed framework is the first to analyze left ventricle wall motion for both global and regional abnormality detection in echocardiography data.

Text

2020-03-04

Attribution-NonCommercial-NoDerivatives 4.0 International

http://hdl.handle.net/2429/73681

Electrical and Computer Engineering

University of British Columbia

UBCV

http://creativecommons.org/licenses/by-nc-nd/4.0/