Electrocardiographic imaging

Abstract

Electrocardiographic imaging (ECGI) is a useful noninvasive modality for exploring the spread of electrical activation within the ventricular wall. In this model study, we seek to determine whether ECGI could detect changes in patterns of pacing-generated epicardial potentials. An anatomical model of the human ventricular myocardium is used to simulate activation sequences initiated at 116 endocardial pacing sites distributed over the left ventricular free wall. From these realistic sequences, we simulate extracardiac potentials at epicardial (202 sites) and torso surfaces (352 sites) using boundary element model of the human torso. ECGI is applied to compute epicardial potentials and unipolar electrograms (202 sites). Inversely computed electrograms correlate well with those simulated by an anatomical model (r>0.9 at 71% of sites). Features of the calculated epicardial potential patterns provide direct visual information about the site of pacing. The results also demonstrate that ECGI provides detailed spatio-temporal reconstruction of patterns of myocardial activation.