Automatic Segmentation and Classification to Diagnose Coronary Artery Disease (AuSC-CAD) Using Angiographic Images: A Novel Framework

Coronary Angiography is necessary because it provides critical information about the blood vessels and blood flow that is essential in diagnosing and treating various cardiovascular conditions. Blood vessel segmentation is a popular approach in medical image analysis since the study of vessels is essential for diagnosis, treatment planning and execution, and evaluation of clinical results. This proposal introduces an Automatic Segmentation and Classification to Diagnose Coronary Artery Disease (AuSC-CAD) framework. Automatic segmentation and classification are key techniques for diagnosing coronary artery disease (CAD) from medical images. The novelty of the proposed framework is the combination of three different levels such as collective constraint-based pre-processing (noise removal, low contrast, and binarization), vessel segmentation, and vessel classification on various modalities of angiography images. Typically, vessel segmentation has been extensively studied, but vessel classification has not received as much attention. The goal of this approach is to extract and analyze relevant features from medical images such as coronary angiograms and CT scans, to diagnose the presence and severity of CAD. The technique involves the segmentation of coronary arteries and classifying the extracted features into normal or abnormal categories. This approach provides a fast, efficient, and accurate way to diagnose CAD, reducing the need for manual interpretation and improving overall diagnostic accuracy. The PSNR (Peak Signal-to-Noise Ratio) values for the pre-processing steps, which include noise removal, contrast enhancement, and binarization, are 34.16, 28.47, and 28.37, respectively.