Automatic segmentation of the carotid artery in ultrasound B-mode images

Abstract

B-mode ultrasound imaging is well-known and used in the medical imaging field; however, it presents various difficulties, specifically in tasks of image segmentation and surface reconstruction, due to intrinsic adverse characteristics, such like low contrast and noise [1,2]. Despite this, B-mode ultrasound imaging has been used in the diagnosis of several cardiac diseases, particularly, carotid artery diseases like atherosclerosis, known as the hardening of the artery, after the accumulation of fatty substances, i.e. lipoproteins, in the artery walls, known as plaques. In this work, an anisotropic diffusion filter is used for speckle removal, and morphological operators are employed in the detection of the artery. The obtained information is used in the definition of two initial contours, one for the lumen and another for the bifurcation boundaries, used to initiate a Chan-Vese-based segmentation method.