gms | German Medical Science

33rd International Congress on Electrocardiology

International Society of Electrocardiology

Inverse Reconstruction Of Activation Wavefronts And Potentials On The Heart Surface Using A Wavefront-Based Model

Meeting Abstract

  • corresponding author presenting/speaker A.A. Ghodrati - Draeger Medical, Danvers, USA
  • D.D.H. Brooks - Northeasten University, Boston, USA
  • G.G. Tadmor - Northeasten University, Boston, USA
  • R.R.S. MacLeod - 3University of Utah, CVRTI, Salt Lake City, USA

33rd International Congress on Electrocardiology. Cologne, 28.06.-01.07.2006. Düsseldorf, Köln: German Medical Science; 2007. Doc06ice061

The electronic version of this article is the complete one and can be found online at:

Published: February 8, 2007

© 2007 Ghodrati et al.
This is an Open Access article distributed under the terms of the Creative Commons Attribution License ( You are free: to Share – to copy, distribute and transmit the work, provided the original author and source are credited.



We describe two methods to solve the inverse problem of electrocardiography using phenomenological models to impose wavefront-like behavior. The first, which we call wavefront-based curve reconstruction (WBCR), reconstructs the activation wavefront modeled as a continuous curve on the heart surface at each time instant. This method uses approximate models of wavefront velocity and the relationship of the wavefront to the heart surface potentials combined with body surface potentials and the torso geometry. In the second method, which we call wavefront-based potential reconstruction (WBPR), we reconstruct the potentials on the heart surface from body surface measurements and a forward model using a Tikhonov regularization approach. However, we include in the Tikhonov regularization an initial estimate, or prior, at each time instant. This estimate is formed from a wavefront-based approximation of the potentials estimated at the previous time instant. Thus the method, in effect, "propagates" a sharp wavefront correction to the smoothed Tikhonov reconstruction. Initial simulation results using measured canine epicardial data with epicardially paced beats show that the WBCR method can capture the anisotropic propagation of the wavefront in the time shortly after pacing, while the WBPR method, even for supraventricular stimulation, simultaneously reconstructs a sharp and accurately-located wavefront and suppresses large errors stemming from the ill-posedness of the inverse problem. Both methods outperform standard Tikhonov approaches and are especially well suited to localizing ectopic or reentrant activity because of their fidelity to the sharp boundaries of activation wavefronts.