Abstract

Three-dimensional (3D) endocardium visualization plays an extremely important role in localization and ablation of target areas and can improve the cure rate in computer-aided cardiac surgery. In this paper, we discuss how to reconstruct and update the corresponding endocardium surface model quickly and accurately based on the sparse point cloud that is collected dynamically on the left atrial intima. Firstly, we construct a specific mesh model, and then collect spatial points, and carried out the mesh grid approximation based on Thin Plate Splines (TPS) intra-operatively, finally, we obtained an approximate target shape after Taubin-based smooth. Experiments on four left atriums' (LA) CT Angiographies (CTA) demonstrated that the reconstructed surfaces can well fit to the target shape, the distance error of surface between the reconstructed and targets can meet clinical requirements, and the deformation time can also meet the need of real-time update. The results showed that our proposed method is robust, fast and flexible to implement.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call