Effect of the torso conductivity heterogeneities on the ECGI inverse problem solution

Zemzemi, Néjib; Dobrzynski, Cécile; Bear, Laura; Potse, Mark; Dallet, Corentin; Coudière, Yves; Dubois, Rémi; Duchâteau, Josselin

doi:10.1109/cic.2015.7408629

Cited by 15 publications

(22 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A single torso model is used in order to reduce potential sources of noise during the comparison of methods. Since, as described in Zemzemi et al ( 2015 ), noise level in clinical practice hides the effect of torso heterogeneity on the inverse solution, this does not imply a loss of generality of the simulation results. These models allow us to simulate different propagation patterns over the atrial surface and the associated BSP (García Mollá et al, 2014 ), from which the epicardial potentials are inversely calculated using different regularization methods.…”

Section: Methodsmentioning

confidence: 92%

Regularization Techniques for ECG Imaging during Atrial Fibrillation: A Computational Study

et al. 2016

View full text Add to dashboard Cite

The inverse problem of electrocardiography is usually analyzed during stationary rhythms. However, the performance of the regularization methods under fibrillatory conditions has not been fully studied. In this work, we assessed different regularization techniques during atrial fibrillation (AF) for estimating four target parameters, namely, epicardial potentials, dominant frequency (DF), phase maps, and singularity point (SP) location. We use a realistic mathematical model of atria and torso anatomy with three different electrical activity patterns (i.e., sinus rhythm, simple AF, and complex AF). Body surface potentials (BSP) were simulated using Boundary Element Method and corrupted with white Gaussian noise of different powers. Noisy BSPs were used to obtain the epicardial potentials on the atrial surface, using 14 different regularization techniques. DF, phase maps, and SP location were computed from estimated epicardial potentials. Inverse solutions were evaluated using a set of performance metrics adapted to each clinical target. For the case of SP location, an assessment methodology based on the spatial mass function of the SP location, and four spatial error metrics was proposed. The role of the regularization parameter for Tikhonov-based methods, and the effect of noise level and imperfections in the knowledge of the transfer matrix were also addressed. Results showed that the Bayes maximum-a-posteriori method clearly outperforms the rest of the techniques but requires a priori information about the epicardial potentials. Among the purely non-invasive techniques, Tikhonov-based methods performed as well as more complex techniques in realistic fibrillatory conditions, with a slight gain between 0.02 and 0.2 in terms of the correlation coefficient. Also, the use of a constant regularization parameter may be advisable since the performance was similar to that obtained with a variable parameter (indeed there was no difference for the zero-order Tikhonov method in complex fibrillatory conditions). Regarding the different targets, DF and SP location estimation were more robust with respect to pattern complexity and noise, and most algorithms provided a reasonable estimation of these parameters, even when the epicardial potentials estimation was inaccurate. Finally, the proposed evaluation procedure and metrics represent a suitable framework for techniques benchmarking and provide useful insights for the clinical practice.

show abstract

Section: Methodsmentioning

confidence: 92%

Regularization Techniques for ECG Imaging during Atrial Fibrillation: A Computational Study

et al. 2016

View full text Add to dashboard Cite

show abstract

“…Nevertheless, the influence of this conductivity on the inverse solution using ventricular signals was not evaluated in the study. Ramanathan concluded in [23] that in clinical application it is not necessary to include torso inhomogeneities in noninvasive electrophysiological imaging but Zemzemi showed in [22] that using inhomogeneities in combination with low noise ECGs improves the accuracy of the inverse solution. These results are in accordance with the results of our study.…”

Section: Discussionmentioning

confidence: 99%

“…Impact of the complexity of the patient-specific geometry (organs and their conductivities in the volume conductor model) on surface potentials was studied by Doessel [21] and Zemzemi [22]. Global study of the influence of the torso inhomogeneities on atrial and ventricular ECG signals was performed in [21].…”

Section: Discussionmentioning

confidence: 99%

Influence of Torso Model Complexity on the Noninvasive Localization of Ectopic Ventricular Activity

Punshchykova

Svehlikova

Tyšler

et al. 2016

Measurement Science Review

View full text Add to dashboard Cite

Location of premature ectopic ventricular activity was assessed noninvasively in five patients using integral body surface potential maps and inverse solution in terms of a single dipole. Precision of the inverse solution was studied using three different torso models: homogeneous torso model, inhomogeneous torso model including lungs and heart ventricles and inhomogeneous torso model including lungs, heart ventricles and atria, aorta and pulmonary artery. More stable results were obtained using the homogeneous model. However, in some patients the location of the resulting dipole representing the focus of ectopic activity was shifted between solutions using the homogeneous and inhomogeneous models. Comparison of solutions with inhomogeneous torso models did not show significantly different dispersions, but localization of the focus was better when a torso model including atria and arteries was used. The obtained results suggest that presented noninvasive localization of the ectopic focus can be used to shorten the time needed for successful ablation and to increase its success rate.

show abstract

“…The classical approach is based on a transfer matrix between epicardial potentials and the torso potentials. To compute this transfer matrix different approaches can be used such as BEM [2] or FEM [13]. However, inverting it is ill-posed [13].…”

Section: Electrocardiographic Imagingmentioning

confidence: 99%

“…To compute this transfer matrix different approaches can be used such as BEM [2] or FEM [13]. However, inverting it is ill-posed [13]. Therefore, different formulations and regularization methods were proposed, see for instance the publications of the ECGI consortium 3 .…”

Section: Electrocardiographic Imagingmentioning

confidence: 99%

Deep Learning Formulation of ECGI for Data-Driven Integration of Spatiotemporal Correlations and Imaging Information

Bacoyannis

Krebs

Cedilnik

et al. 2019

Lecture Notes in Computer Science

View full text Add to dashboard Cite

The challenge of non-invasive Electrocardiographic Imaging (ECGI) is to recreate the electrical activity of the heart using body surface potentials. Specifically, there are numerical difficulties due to the ill-posed nature of the problem. We propose a novel method based on Conditional Variational Autoencoders using Deep generative Neural Networks to overcome this challenge. By conditioning the electrical activity on heart shape and electrical potentials, our model is able to generate activation maps with good accuracy on simulated data (mean square error, MSE = 0.095). This method differs from other formulations because it naturally takes into account spatio-temporal correlations as well as the imaging substrate through convolutions and conditioning. We believe these features can help improving ECGI results.

show abstract

Effect of the torso conductivity heterogeneities on the ECGI inverse problem solution

Cited by 15 publications

References 11 publications

Regularization Techniques for ECG Imaging during Atrial Fibrillation: A Computational Study

Regularization Techniques for ECG Imaging during Atrial Fibrillation: A Computational Study

Influence of Torso Model Complexity on the Noninvasive Localization of Ectopic Ventricular Activity

Deep Learning Formulation of ECGI for Data-Driven Integration of Spatiotemporal Correlations and Imaging Information

Contact Info

Product

Resources

About