Detection of Coronary Artery Blockage at an Early Stage using Effective Deep Learning Technique

Coronary artery blockage is the most common and frequent disease affecting the human body. Our proposed method aims toward a fully automated system for detecting a coronary artery blockage at an early stage using image processing and deep learning technique that can improve medical treatment of the heart. This research aims to implement a system that can detect stenosis areas of the coronary artery due to the buildup of cholesterol, plaque, and other blocking agents using 2D angiogram images. Evaluating 2,151 angiogram image datasets, we trained and tested our models to reach a conclusion. The research uses CNN model architecture that uses a dataset of 2D angiogram images of the segmented coronary arteries, which are analyzed using VGG16, VGG19, ResNet50, InceptionV3, and DenseNet121 models. For further enhanced results, we classified our dataset in two ways; Binary Classification and Multi-class Classification. Next, using the Ensemble model architecture, we evaluated the results and accuracy of the models. According to our Confusion Matrix evaluation, our Binary Class Ensemble model detected 85 blocks and 557 no-block images. Additionally, for the Multi-class Ensemble Model, the algorithm successfully classified 10 LAD, 7 LCX, 21 RCA, 42 Multiple block images, and 557 no-block images. Finally, we achieved accuracy, precision, recall, and an F1 Score greater than 0.99 for the Binary Classification and greater than 0.98 for the Multi-class classification. In this way, deep learning techniques can improve and develop medical science at a prodigious level resulting in error-free medical treatment of the heart at an early stage.