A review of computational methods applied for identification and quantification of atherosclerotic plaques in images

A review of methodologies developed to analyze atherosclerotic plaques in images is presented.The traditional visual analysis of atherosclerotic plaques in images is introduced.The segmentation of atherosclerotic plaques in ultrasound, CT and MR images is addressed.Techniques of image processing, clustering and classification are reviewed.The effectiveness and drawbacks of each technique are discussed. Evaluation of the composition of atherosclerotic plaques in images is an important task to determine their pathophysiology. Visual analysis is still as the most basic and often approach to determine the morphology of the atherosclerotic plaques. In addition, computer-aided methods have also been developed for identification of features such as echogenicity, texture and surface in such plaques. In this article, a review of the most important methodologies that have been developed to identify the main components of atherosclerotic plaques in images is presented. Hence, computational algorithms that take into consideration the analysis of the plaques echogenicity, image processing techniques, clustering algorithms and supervised classification used for segmentation, i.e. identification, of the atherosclerotic plaque components in ultrasound, computerized tomography and magnetic resonance images are introduced. The main contribution of this paper is to provide a categorization of the most important studies related to the segmentation of atherosclerotic plaques and its components in images acquired by the most used imaging modalities. In addition, the effectiveness and drawbacks of each methodology as well as future researches concerning the segmentation and classification of the atherosclerotic lesions are also discussed.