Blood Vessels Segmentation of Coronary X-Rays Angiography Images Including Edge based Features and Artificial Intelligence Approaches

Abstract

Abstract In present Era, the cardiovascular disease is the most common disease in human. According to the World Health organization reports 2022, there are 70% of Human death from the Heart attack. Most of the Indian peoples suffering from heart disease having the age group of 30–60 years. Xray Coronary angiography imaging is a primary procedure for diagnosis of heart disease. Manual Segmentation of heart vessels by cardiologists are typical and time-consuming process. Manual segmentation facing the problem of variations in results due to experience and expertise of the medical professionals. Segmentation of coronary vessels angiography provides important information for the expert and patient suffering from cardiovascular disease. Therefore, different types of computer-aided Tools have been designed and developed for automatic segmentation of coronary vessels angiography images. An automatic segmentation of coronary arteries can be improved by computer vision and artificial intelligence approaches. In this paper an automatic segmentation of coronary angiography images has been designed and implemented using edge-based feature and artificial intelligence approaches. For this purpose, dominating and prominent edges of cardiovascular arteries system has been detected using traditional edge detection algorithms like Sobel, Prewitt, Robert’s and Canny. The strong edges from the above-mentioned algorithms are selected using Artificial Intelligence (Random Forest) algorithm. Experimental results shows that proposed model provides accuracy, Positive Prediction Value, Sensitivity and Dice Coefficient as 99%, 96%, 94% and 95% respectively.