Publikation

Towards learning-based catheter distal section steering

Bingbin Yu, Abraham Temesgen Tibebu, Yohannes Kassahun, Felix Bernhard, Emmanuel Vander Poorten

In: 4th Joint Workshop on New Technologies for Computer/Robot Assisted Surgery. Joint Workshop on New Technologies for Computer/Robot Assisted Surgery (CRAS-2014), October 14-16, Genoa, Italy. Joint Workshop on New Technologies for Computer/Robot Assisted Surgery (CRAS-2014) 4th October 14-16 Genoa Italy Seiten 167-170 10/2014.

Abstrakt

Despite the growing popularity of catheter based therapy in cardiovascular diseases, the operation of steerable catheters is still inaccurate due to their inherent compliance and in case of cable-based steerable catheters friction between cable and catheter sleeve. The performance of most catheter therapies such as cardiac ablation depends on the accuracy of catheter tip navigating and settling on the target position of the heart tissue, which is moving dynamically when the heart is still beating. Since it is a tedious task, in fact any method to improve the catheter tip steering is welcomed. To improve the positioning accuracy of cable-actuated steerable catheters, this paper proposes the use of joint probability density [1] based catheter modelling and model based catheter steering. The joint probability distribution of the catheter handle displacement, the catheter tip position, the catheter shape and the bending angle is learned. Based on the learned model, a position control of a conventional ablation catheter is developed.

Deutsches Forschungszentrum für Künstliche Intelligenz
German Research Center for Artificial Intelligence