C Ramanathan1, Y Rudy. 1. Cardiac Bioelectricity Research and Training Center, Department of Biomedical Engineering, Case Western Reserve University, Cleveland, Ohio 44106-7207, USA.
Abstract
INTRODUCTION: Noninvasive electrocardiographic imaging (ECGI) involves inverse reconstruction of epicardial potentials, electrograms (EGMs), and isochrones from body surface potential maps (BSPMs). The heart lies in a volume conductor that includes lungs, blood, bone, muscle, and fluid. We investigate the effects of these torso inhomogeneities on reconstructed epicardial potentials, EGMs, and isochrones to address the issue of whether they should be included in clinical ECGI methodology. METHODS AND RESULTS: Potential data were obtained for different pacing protocols from a dog heart suspended in a human-shaped torso tank. Accurate geometry of torso inhomogeneities was digitized from the Visual Human Project and appropriately introduced into a computer model of the torso. Three models were used: accurate inhomogeneous torso, homogeneous torso, and a torso with stylized lungs (to generate an approximate model). The inhomogeneous model was used to compute BSPMs from the measured epicardial potentials. These BSPMs were the starting point for inverse computations in the different torso models. Epicardial potential maps, EGMs, and isochrones were computed. The homogeneous model produced slightly less accurate epicardial potential reconstructions than the inhomogeneous model and stylized lung model, but epicardial potential patterns, EGMs, isochrones, and locations of pacing sites were reconstructed with comparable accuracy when torso inhomogeneities were ignored. CONCLUSION: The results demonstrate that, in the clinical application, it is not necessary to include torso inhomogeneities for noninvasive reconstructions of epicardial potentials, EGMs, and activation sequences.
INTRODUCTION: Noninvasive electrocardiographic imaging (ECGI) involves inverse reconstruction of epicardial potentials, electrograms (EGMs), and isochrones from body surface potential maps (BSPMs). The heart lies in a volume conductor that includes lungs, blood, bone, muscle, and fluid. We investigate the effects of these torso inhomogeneities on reconstructed epicardial potentials, EGMs, and isochrones to address the issue of whether they should be included in clinical ECGI methodology. METHODS AND RESULTS: Potential data were obtained for different pacing protocols from a dog heart suspended in a human-shaped torso tank. Accurate geometry of torso inhomogeneities was digitized from the Visual Human Project and appropriately introduced into a computer model of the torso. Three models were used: accurate inhomogeneous torso, homogeneous torso, and a torso with stylized lungs (to generate an approximate model). The inhomogeneous model was used to compute BSPMs from the measured epicardial potentials. These BSPMs were the starting point for inverse computations in the different torso models. Epicardial potential maps, EGMs, and isochrones were computed. The homogeneous model produced slightly less accurate epicardial potential reconstructions than the inhomogeneous model and stylized lung model, but epicardial potential patterns, EGMs, isochrones, and locations of pacing sites were reconstructed with comparable accuracy when torso inhomogeneities were ignored. CONCLUSION: The results demonstrate that, in the clinical application, it is not necessary to include torso inhomogeneities for noninvasive reconstructions of epicardial potentials, EGMs, and activation sequences.
Authors: Miguel Rodrigo; Andreu M Climent; Alejandro Liberos; Francisco Fernández-Avilés; Omer Berenfeld; Felipe Atienza; Maria S Guillem Journal: Heart Rhythm Date: 2017-04-10 Impact factor: 6.343
Authors: Azar Rahimi; John Sapp; Jingjia Xu; Peter Bajorski; Milan Horacek; Linwei Wang Journal: IEEE Trans Med Imaging Date: 2015-08-04 Impact factor: 10.048