Title: Left Ventricular Deformation Analysis from 4D Echocardiography: A Machine Learning Approach
Abstract: Myocardial infarction (MI) remains a leading cause of morbidity and death in many countries. Acute MI causes regional dysfunction, which places remote areas of the heart at a mechanical disadvantage resulting in long term adverse left ventricular (LV) remodeling and complicating congestive heart failure (CHF). Echocardiography is a clinically established, cost-effective technique for detecting and characterizing coronary artery disease and myocardial injury by imaging the left ventricle (LV) of the heart. In this talk, an image analysis system to derive quantitative 4D (three spatial dimensions + time) echocardiographic (4DE) deformation measures (i.e. LV strain) for use in diagnosis and therapy planning will be described. This system combines displacement information from shape tracking of myocardial boundaries (derived from B-mode echocardiographic data) with mid-wall displacements from radio-frequency-based ultrasound speckle tracking to estimate myocardial strain. The talk will first overview our earlier efforts based on Bayesian analysis and radial basis functions for integrating information. Next, a new robust approach for estimating improved dense displacement measures based on an innovative data-driven, deep feed-forward, neural network architecture that employs domain adaptation between data from labeled, carefully-constructed synthetic models of physiology and echocardiographic image formation (i.e. with ground truth), and data from unlabeled noisy in vivo porcine or human echocardiography (missing or very limited ground truth) will be discussed. Test results on LV strain will be presented from synthetic and in vivo 4DE image sequence data, including a comparison to strains derived from MR tagging. Finally, planned applications to rest-stress imaging and image-guided hydrogel therapy will also be described.
James Duncan, the Ebenezer K. Hunt Professor of Biomedical Engineering, has focused his research and teaching in the areas of biomedical image processing and analysis. He holds joint appointments in diagnostic radiology and electrical engineering, is the associate chair and director of undergraduate studies in the Department of Biomedical Engineering as well as the vice-chair for bioimaging sciences research in diagnostic radiology. Duncan is a fellow of the Institute of Electrical and Electronics Engineers (IEEE) and the American Institute for Medical and Biological Engineering. He is president of the International Society for Medical Image Computing and Computer Assisted Intervention and is a member of the American Association for Artificial Intelligence and the I.E.E.E. Computer Society, among other professional organizations.