MEDICAL IMAGING AND PROCESSING METHODS FOR CARDIAC FLOW RECONSTRUCTION

Abstract

Intra-cardiac blood flow imaging and visualization is challenging due to the processes involved in generating velocity fields of flow within specific chambers of interest. Visual analysis of cardiac flow or wall deformation is crucial for an accurate examination of the heart. Cardiac chamber boundary encapsulation is one of the key implementations for region definition. To provide intelligible results describing flow within the human heart, cardiac chamber segmentation is a pre-requisite so that fluid motion information can be presented within a region of interest defined by the chamber boundary. A technique that is used to establish contouring along the cardiac wall is described mathematically. This article also sets the practical foundation for flow vector synthesis and visualization in the cardiac discipline. We have outlined conceptual development and the construction of flow field based on a three-dimensional Cartesian grid that can give a greater insight into the blood dynamics within the heart. We developed a framework that is able to present both anatomical as well as flow information by overlaying velocity fields over medical images and displaying them in cine-mode. By addressing most of the methods involved from the programming perspective, procedural execution and memory efficiency have been considered. Our implemented system can be used to examine abnormal blood motion behaviour or discover flow phenomena in normal or defective hearts.