A novel vascular centerline extraction method of CT angiography

Objective To extract the vascular centerline of CT angiography. Methods In order to obtain centerline branch information, a method based on minimum path and Bayesian tracking was proposed in this article. The algorithm in this article was divided into two steps: firstly, a new multiscale filtering algorithm was used to enhance vascular structure; secondly, vascular centerline extraction algorithm based on minimum path and Bayesian tracking was performed. Result The proposed method not only could effectively extract vascular centerline, obtain centerline branch information. Conclusion Compared with similar algorithms, this algorithm can enhance the vascular structure well and inhibit the noise and artifacts effectively. Key words: Vascular centerline; Centerline branch information; Minimal path; Bayesian tracking