Noninvasive Localization of Premature Ventricular Activity Using Different Equivalent Point Sources

Abstract

IntroductionOne of the aims of electrocardiographic imaging methods (inverse methods in ECG) is the noninvasive identification of the origin of a premature ventricular contraction (PVC) [1]. Reliable results could be helpful for planning the ablation procedure used for the treatment of such heart disease. Calculation of the inverse results depends on the type of equivalent heart generator used in mathematical formulation of the problem. The aim of this study was to compare the single point inverse solutions assuming three types of a heart generator from the measured repeated PVCs in one patient. Materials and methodsMeasured body surface potential maps (BSPMs) were used as the input data for the inverse solution. The ECG signals were recorded in 63 precordial leads during nine spontaneous ectopic beats (PVCs) in one patient undergoing the ablation procedure [9] (patient 20). The measured dataset was provided in EDGAR database by the Institute of Biomedical Engineering, Karlsruhe Institute of Technology (KIT), Germany and the First Department of Medicine (Cardiology), University Medical Centre Mannheim, Germany. From the ablation procedure two reference points indicating possible origins of the spontaneous PVC were assigned with the CARTO system: PVC1 -the site where the earliest activation time was recorded and PVC2 -the site of the latest successful ablation.The patient-specific geometry was obtained from MR scans taken in the end-diastolic phase and described by surface triangular meshes of the geometry of the torso and heart ventricles.In EDGAR database the precomputed forward transfer matrices computed by boundary element method for two types of equivalent heart generators were provided: epicardial potentials (EPs) and transmembrane voltages (TMVs). The positions of the generators were assumed in the vertices of the joined triangulated endo and epicardial surface. The transfer matrix for dipoles (DIPs) situated in the same vertices was also computed.The single point inverse solution for localization of the