A Method for Registration of 3-D Shapes
IEEE Transactions on Pattern Analysis and Machine Intelligence - Special issue on interpretation of 3-D scenes—part II
A Stochastic Iterative Closest Point Algorithm (stochastICP)
MICCAI '01 Proceedings of the 4th International Conference on Medical Image Computing and Computer-Assisted Intervention
Hi-index | 0.00 |
To improve the intra-operative image fusion performance in the ablation procedure of atrial fibrillation (AF) treatment, this paper presents a novel registration method for CT segmented surfaces and three-dimensional (3-D) cardiac electroanatomical maps. Random perturbation is introduced to deform the electroanatomical maps in the registration process. The magnitude of deformation automatically attenuates during iterations. Compared to the typical iterative closest point (ICP) algorithm that often converges to local minima, the proposed algorithm is much less sensitive to the initial transformation and able to move out of the local minima and converge to a solution with smaller registration error. Through experiments using both in vivo and simulation data, the results show significant improvements on the registration accuracy and success rate over the existing method being used in the clinical environment. The improved intra-operative registration results can help physicians easily navigate the catheter during the AF interventional procedures.