Canine left ventricular purkinje fiber network construction using manifold learning

Purkinje fiber network (PFN), one of the most important components of the ventricular conduction system, is crucial in modeling ventricular tachycardia and fibrillation. Construction of anatomical detail Purkinje fiber network, however, is a very challenging task. In this paper, we present a novel method for restoring the 3D PFN in the left ventricle (LV) by manifold learning. Motivated by the fact that canine Purkinje fiber is generally on the endocardial surface of the heart, we have collected a set of real 2D canine LV images, from which the PFN image is detected and segmented. We then use manifold learning to map 3D canine left ventricular model to 2D PFN and the inverse mapping to finish the final construction. Our experimental results show that the 3D PFN construction method is flexible and feasible.