Automatic 3D Aorta Segmentation in CT Images

Cardiovascular disease is one of the most common high incidence diseases, which leads to the urgent demand of 3D aorta shape reconstruction with CT images for help doctors making effective and accurate diagnosis. In literature, 3D aorta reconstruction methods with CT images mainly were based manual or semiautomatic operations, which would limit the practical applications of the technique. In the paper, a fully automatic 3D aorta segmentation algorithm was proposed. Firstly, the 2D shape of aortic arch in a CT image was utilized to locate the 3D position in CT image sequence. Then, a level set method was adopted to segment all the aortic edge in the whole CT image sequence by taking the aortic arch edge as the initial contour. Finally, the algorithm was testified on CT image dataset, and the experimental results show that our method can reconstruct the 3D shape with CT image full automatically.