Automated coronary artery analysis system in 3D CTA images

This paper presents an automated computer assistant system for segmentation and quantification of the coronary arteries in 3D contrast enhanced computed tomography angiography (CTA). To the best of our knowledge, no commercially available software package permits fully automated detection and assessment of atherosclerotic stenosis so far. Therefore, in clinical practice, the evaluation of the coronary CT studies is accomplished by radiologists, which is obviously a tedious and time consuming task, since it usually involves hundreds of axial cross sectional CT images to be examined. In the proposed system, the main branches of the coronary arteries are firstly extracted from the volume datasets by employing a regional statistic based active contours method. Next, calcified voxels are detected in the within the coronary vessels and removed. The centreline of the coronaries is extracted by mesh contraction algorithm in the following step. Finally, we compute the circularity and the convex hull deficiency of the cross sections at specific locations along the coronary arteries, which are then used to automatically detect and identify the presence of calcified stenosis.