Quantitative analysis of three-dimensional left ventricular wall motion for diagnosis of ischemia and infarction

Cardiologists have sought measures of left ventricular wall motion that could be used to improve the accuracy of echocardiographic tests to screen for coronary artery disease. Recently, a new ultrasound technology called real-time three-dimensional echocardiography (RT3DE) emerged enabling new solutions to problems that previously prevented development of robust wall motion measures. Quantitative analysis of cardiac images should provide a more objective and accurate approach to diagnosis than visual analysis that is typically used. The main objective of this dissertation was to develop a novel quantitative wall motion analysis capable of detecting ischemia (insufficient blood supply for the demand in the tissue) and infarction (tissue death) in 3D cardiac images.;One of the most important results of the analysis is a novel measure of wall motion, three-dimensional fractional shortening (3DFS). 3DFS is derived by fitting endocardial surface data in prolate spheroidal coordinates and provides a measure of the fractional change in 3D position of the left ventricular endocardium from end diastole to end systole. In this dissertation, 3DFS was initially tested on computational simulations of ischemia where it predicted the size of simulated ischemic regions. The computer simulations also identified a fundamental limit in the ability to predict the size of very small ischemic regions from wall motion analysis.;A validation study in animals was performed to assess the ability of 3DFS to predict the size and location of acutely ischemic regions. Regions of abnormal wall motion were identified in RT3DE images by 3DFS and by two cardiologists. Regions of abnormal wall motion were compared to colored microsphere measurements, an accurate method for assessing myocardial perfusion. 3DFS localized ischemic regions as accurately as cardiologists' visual scoring, while providing superior quantification of ischemic region size.;To prepare the wall motion analysis for clinical application, 3DFS was measured from RT3DE and magnetic resonance images in healthy volunteers providing a database of the range of normal levels. Finally, new wall motion analysis methods were developed that differentiated between acutely ischemic and healed infarct regions in computational simulations.