Left atria reconstruction from a series of sparse catheter paths using neural networks

Abstract

Modeling and reconstructing the shape of a heart chamber from partial or noisy data is useful in many (minimally) invasive heart procedures. We propose a method to reconstruct the shape of the left atria during the electrophysiology procedure from a series of simple catheter maneuvers. We use left atria shapes generated from a statistical based physical model and approximate traversal locations of catheter maneuvers inside the left atria. These paths mimic realistic ones doable in a lab phantom. We demonstrate the ability of a deep neural network to approximate the atria shape solely based on the given paths. We compare the results against training from partial data generated by the intersection of a randomly generated sphere and the atria. We test the presented network on actual lab phantoms and show promising results.