An Automatic Segmentation Algorithm of Boundary Layers from B-Mode Ultrasound Images

Abstract

In image segmentation technique, partition an image into meaningful regions with respect to a particular application. In a digital image, pixels of a region have some common calculated properties like color, intensity, texture, depth, motion, etc. These properties are used to cluster pixels of a region in the given image. Segmented images are more meaningful, easy to analyze and are used to locate objects and boundaries. Segmentation of an image is the preliminary step to pattern recognition. Here, we propose an automatic segmentation method of the boundary layers for calculating the value of Intima-Media Thickness (IMT) from lumen-intima & media-adventitia layers without any user intervention from the B-mode ultrasound image of the common carotid artery. The main objective of our work is to explore the possibility and feasibility of a computational approach for the automatic segmentation of boundary layers.