Coronary Artery Blockage Identification and Severity Classification using a hybrid convolutional neural network (CNN) model

Abstract

Abstract

    The cardiac diseases are considered a major health problem. Therefore the early identification and detection of the heart issues is an important procedure. Technology plays a significant role in this health issue.  It is an ideal means that helps experts to diagnose the heart problems. The Coronary Artery Analysis is a medical heart test that checks the coronary arteries. The medical imaging techniques of  the MRI, and CT scans are used to identify abnormalities such as blockages, narrowing, or other structural issues. This analysis helps assess the severity of coronary artery disease (CAD) and guides treatment decisions to improve heart health. This research presents a hybrid convolutional neural network (CNN) model integrated with sophisticated optimization approaches for the analysis of heart/MRI ultrasound images. The research utilizes a hybrid approach of Gradient Boosting (GB) and Support Vector Machine (SVM) classifiers to detect coronary artery abnormalities. The suggested methodology precisely identifies Coronary Artery Blockages (CAB) in medical imaging and delineates regions of concerns. In the next phase, a tailored CNN classifier is employed to quantify the dimensions of obstructions, encompassing their length and severity, so facilitating accurate anomaly evaluation for diagnostic assistance in healthcare environments.