Segmentation and visualization of left ventricle in MR Cardiac images

Abstract

This thesis developes a segmentation algorithm to extract endocardial and epicardial countours of left ventricle in MRI. The algorithm is based on the gradient-vector-flow (GVF) snake and the prediction of initial contour (PIC). Especially, the proposed algorithm uses physical characteristics of endocardial and epicardial contours, cross profile correlation matching (CPCM), and mixed interpolation model. In the experiment, we used short axis volumetric image sets, which were obtained by Siemens, Medinus, and GE MRI Systems. The experimental results show that the proposed algorithm can extract acceptable epicardial and endocardial walls. We calculate quantitative parameters using segmented results, and show these graphically. The segmented left ventricle is visualized volumetrically by surface rendering. The proposed algorithm is implemented on Windows environment using Visual C++.