Ali K. Hamou, September 2001--June 2003, "Segmentation of Carotid Artery Ultrasound Images", degree conferred on Autumn 2003 convocation, Computer Science Department, University of Western Ontario, Canada.
M.Sc. Thesis Abstract
Over the past few years, medical imaging techniques have been evolving to keep up with the most current technological developments. A major advance was the arrival of ultrasound imaging and its uses. Ultrasound provides a non-invasive means for visualizing various tissues within the human body. However, these visualizations tend to be filled with speckle noise and other artifacts, due to the nature of sound waves.
This thesis presents a novel segmentation technique for use on noisy B-mode ultrasound images of the carotid artery. This scheme is based on several commonly used image-processing techniques, such as histogram equalization, Canny edge detection and morphology. The proposed scheme provides various degrees of customizability, for a wide range of ultrasound images. The experimental results show that this scheme accurately segments the different textures in ultrasound images. These segmented regions alleviate the need for a technician's manual segmentation of wanted regions. Moreover, this technique can be used in conjunction with other schemes to uncover further details of the carotid artery, (such as plaque levels and flow).