Desarrollo de un aplicativo de software para la segmentación de arterias coronarias a partir de imágenes angiograficas, utilizando el metodo de crecimiento de regiones.

Abstract

In this work the development of an automated system that allows segmentation of the coronary arteries on angiography images arises. The methodology to be implemented is based on two main phases: the pre-processing and processing. The first aims to improve the quality of the images and delete unwanted objects in these (ribs, spine, lungs, image guidance, etc.), this process is carried out using techniques such as logarithmic subtraction, contrast enhancement algorithms, mathematical morphology, among others. The processing phase is based on an algorithm of region growing, the core of this work, where an initial pixel called seed, located in the area of interest is compared with neighboring pixels, studying features like intensity, topological relationship and characterization regions. From this image are automatically created new seeds to segment the remaining 2-D images. The database used was obtained at the University Hospital of the University of Los Andes (IAHULA), Merida, Venezuela Autonomous Institute. A level of performance you can appreciate, in the 3 studies used a good segmentation of the coronary tree, thanks to a good background blurring is generated and many of the factors that affect this type of images are deleted, this product the achievement of selected techniques. Similarly, a collection of images corresponding to the user interface is presented as a final product. Finally, it is important to note that this image catheter could be removed almost entirely, compared with other studies. [20]