Automatic lumen segmentation in CT and PC-MR images of abdominal aortic aneurysm

Vascular segmentation through the use of image processing tools provides significant information that allows for the accurate diagnosis, categorization, registration, and visualization of vascular disease. Currently, in the assessment of Abdominal Aortic Aneurysms (AAA), radiologists manually segment different regions on interest on each medical image to create a full volume of the abdominal aorta. Such manual segmentation is a time consuming task, prone to errors and a subjective approach especially when non-contrast enhanced images are present. In this paper, we introduce an automatic system to segment the aortic lumen in non-contrast enhanced CT scans and PC-MR images using digital image processing algorithms where image enhancement, denoising, edge detection, and regional growing algorithms are utilized. The output of this work forms the basis for a future reliable inner and outer wall segmentation of the AAA.