Characterization of Atheroma Plaques Echogenicity and Texture for Computer- Aided Diagnosis in Cardiovascular Diseases

Abstract

Developments of image processing techniques such as classical feature extraction, neural networks, artificial intelligence and deep learning have made possible enhancement, analysis, recognition and classification of medical imaging. The presence of atheroma plaques in the carotid artery is associated with an increased risk of all forms of cardiovascular disease and this risk increases with plaque growth. Moreover, the morphological structure of the plaque is closely related to the overall cardiovascular risk. Regardless of the type of medical imaging computer-aided diagnosis (CAD) has become one of the major research subjects in medical imaging and diagnostic. The purpose of this article is to obtain a fully automatic procedure for atheroma plaques characterization in order to prevent stroke. Our approach is based on two morphological characteristics of atheroma plaques: the measurement of echogenicity based on percentage of white and the quantification of homogeneity based on five Haralick texture features.