Performance analysis of extracting blood vessel blockages from coronary angiogram images using soft computing classification techniques

Abstract

In the approach of new medical imaging technologies, it is possible to explore the anatomical and functional structure of the human heart. Coronary Artery Disease (CAD) is a significant cause of death worldwide. There are numerous tests available for recognition of CAD at present. Quantitative Coronary Angiography (QCA) is the highest level imaging strategy for diagnosing CAD and evaluating the level of stenos is. There are many causes for the heart disease. One of the primary reason is, due to the narrow blood vessel blockage of an artery. Angiography is a specific X-ray imaging technique minimally non invasive procedure used to diagnose coronary artery disease. Extracting the maximum possible information from an image obtained is essential. However, image processing models based on scan sections or radiographic views may not wholly provide diagnostic details at an early stage. These limitations necessitate the development of new analysis techniques that will improve diagnostic ability. The main aim of the thesis is connected with the removal of artifacts and measurement of the length of coronary artery vessel blockage in X-ray coronary angiography images which makes an essential and fundamental component of a computer-aided system intended for the early identification of CAD. This thesis aims to explore the methods such as pre-processing, enhancement, segmentation, a classification of coronary artery vessel blockage and non-vessel blockage in the coronary artery or the assessment of the length of the vessel blockage newline