Summary: Internal carotid arteries, which carry oxygen-rich blood to the brain, are very vital for life. Congestions and thinning in these arteries can lead to serious damage in the human body that may even result in death. Carotid artery stenosis is usually narrowing caused by atherosclerosis or lumen bottleneck of the carotid artery. Carotid artery stenosis is usually narrowing caused by atherosclerosis or lumen bottleneck of the carotid artery. Carotid arteries show close proximity to bone and bony structures. Bone tissue and carotid arteries are often confused with each other when vessel evaluations perform. For this reason, it is usually a doubt whether the potential vessel disruption continues with veins or other structures. The main issue when trying to segment vessels full with contrast agent is the density overlap between vessel and bone. The purpose of this thesis is to make decision support system about diagnosis of the disease by extracting the carotid arteries from computerized tomography angiography images. In this study, bone and cartilage tissues are separated by using the inverse method from CTA images with only one scanning. Then, vessel segmentation is performed and the results are visualized in 3D.
Key Words: Carotid arteries, blood vessels, vessel segmentation, image processing,
machine learning, 3D segmentation, visualization |