Geodesic Distance Algorithm for Extracting the Ascending Aorta from 3D CT Images

Figure 2

The typical good, moderate, and poor CCTA images used in the evaluation. These images were graded according to noise and artifacts such as step artifacts due to wrong registration and beam hardening. (a) and (d) show images of good quality with little noise and no step artifact (data#000). (b) and (e) show images of moderate quality with some noise and minor step artifact (data#003). (c) and (f) show images of poor quality with noise and major step artifact (data#008). The images on the top are axial plane ((a),(b),(c)) and the images at the bottom are coronal plane((d),(e),(f)).

(a)

(b)

(c)

(d)

(e)

(f)