Committee:

Dr. Yezzi, Advisor

Dr. Vela, Chair

Dr. Oshinski

Dr. Dixon

Abstract: The objective of the proposed research is to develop a novel multi-dimensional automatic edge detection algorithm based on shape priors and principal component analysis (PCA) for automatic segmentation of myocardial boundaries in cardiac CT imagery. Function and anatomy are both equally important in the assessment of coronary artery disease (CAD). Currently, there are few, if any, tools that can perform full automatic cardiac segmentation available in the market, which hampers the translation of a multi-modality fusion approach to the clinical environment. We propose to develop algorithms which will allow efficient segmentation of the left and right endocardia and biventricular epicardium from clinical CT datasets with minimal user intervention, clinically acceptable times and accuracy.