Estimation of Local Conduction Velocity from Myocardium Activation Time: Application to Cardiac Resynchronization Therapy

Pheiffer, Thomas S.; Soto‐Iglesias, David; Nikulin, Yaroslav; Passerini, Tiziano; Krebs, Julian; Sitges, Marta; Berruezo, Antonio; Cámara, Óscar; Mansi, Tommaso

doi:10.1007/978-3-319-59448-4_23

Cited by 6 publications

(10 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Since ground truth intramural activation times are commonly not available from in-vivo interventions, this work relies on a synthetic dataset generated by a fast graph-based electrophysiological model (Pheiffer et al, 2017 ). The physiological priors of the cardiac anatomy are expected to be known.…”

Section: Methodsmentioning

confidence: 99%

“…To this end, we require a personalization scheme that finds the best set of edge weights that explains the data, i.e., the sparse electroanatomical map as well as the routinely acquired 12-lead ECG. This work leverages the approach proposed by Pheiffer et al ( 2017 ). The first step comprises a global optimization of homogeneous tissue conduction velocities.…”

Section: Methodsmentioning

confidence: 99%

“…A second step refines the edge weights to match the sparse measurements (Pheiffer et al, 2017 ). The objective function in this case is the mean squared error loss over the M vertices with measurements…”

Section: Methodsmentioning

confidence: 99%

“…In addition, the model requires a robust and accurate personalization scheme to estimate local tissue properties from measurements. For instance, Pheiffer et al proposed a personalization scheme of local tissue conductivities from a left-ventricular endocardial EAM using backpropagation of errors along the wave propagation paths and gradient descent to tune the tissue properties (Pheiffer et al, 2017 ). Even though standard personalization methods are able to match the measured data, their accuracy is dependent on a careful selection of the modeling parameters and the boundary conditions, e.g., the choice of electrical propagation origin.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Extrapolation of Ventricular Activation Times From Sparse Electroanatomical Data Using Graph Convolutional Neural Networks

et al. 2021

Self Cite

View full text Add to dashboard Cite

Electroanatomic mapping is the gold standard for the assessment of ventricular tachycardia. Acquiring high resolution electroanatomic maps is technically challenging and may require interpolation methods to obtain dense measurements. These methods, however, cannot recover activation times in the entire biventricular domain. This work investigates the use of graph convolutional neural networks to estimate biventricular activation times from sparse measurements. Our method is trained on more than 15,000 synthetic examples of realistic ventricular depolarization patterns generated by a computational electrophysiology model. Using geometries sampled from a statistical shape model of biventricular anatomy, diverse wave dynamics are induced by randomly sampling scar and border zone distributions, locations of initial activation, and tissue conduction velocities. Once trained, the method accurately reconstructs biventricular activation times in left-out synthetic simulations with a mean absolute error of 3.9 ms ± 4.2 ms at a sampling density of one measurement sample per cm2. The total activation time is matched with a mean error of 1.4 ms ± 1.4 ms. A significant decrease in errors is observed in all heart zones with an increased number of samples. Without re-training, the network is further evaluated on two datasets: (1) an in-house dataset comprising four ischemic porcine hearts with dense endocardial activation maps; (2) the CRT-EPIGGY19 challenge data comprising endo- and epicardial measurements of 5 infarcted and 6 non-infarcted swines. In both setups the neural network recovers biventricular activation times with a mean absolute error of less than 10 ms even when providing only a subset of endocardial measurements as input. Furthermore, we present a simple approach to suggest new measurement locations in real-time based on the estimated uncertainty of the graph network predictions. The model-guided selection of measurement locations allows to reduce by 40% the number of measurements required in a random sampling strategy, while achieving the same prediction error. In all the tested scenarios, the proposed approach estimates biventricular activation times with comparable or better performance than a personalized computational model and significant runtime advantages.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Extrapolation of Ventricular Activation Times From Sparse Electroanatomical Data Using Graph Convolutional Neural Networks

et al. 2021

Self Cite

View full text Add to dashboard Cite

show abstract

“…The possibility of integrating this source of information with activation time data would improve the accuracy of conduction velocity estimates, as well as the integration of additional data on the position and the shape of the mapping catheter (see e.g., Verma et al, 2018 ). We remark that our model parameterization, presented in the next section, can take advantage of any conduction velocity estimation technique, starting from the most adopted ones reviewed in Cantwell et al ( 2015 ), up to new techniques, such as the streamline-based method of Good et al ( 2020 ), the two-stage technique based on the depolarization pattern reconstruction shown in Nagel et al ( 2019 ), the back-propagation parameter estimation procedure proposed in Pheiffer et al ( 2017 ), or physics-informed neural networks applied to cardiac activation mapping in Sahli Costabal et al ( 2020 ).…”

Section: Estimation Of the Conduction Velocitymentioning

confidence: 99%

A Computational Study of the Electrophysiological Substrate in Patients Suffering From Atrial Fibrillation

Pagani

Dedè

Frontera³

et al. 2021

Front. Physiol.

View full text Add to dashboard Cite

In the context of cardiac electrophysiology, we propose a novel computational approach to highlight and explain the long-debated mechanisms behind atrial fibrillation (AF) and to reliably numerically predict its induction and sustainment. A key role is played, in this respect, by a new way of setting a parametrization of electrophysiological mathematical models based on conduction velocities; these latter are estimated from high-density mapping data, which provide a detailed characterization of patients' electrophysiological substrate during sinus rhythm. We integrate numerically approximated conduction velocities into a mathematical model consisting of a coupled system of partial and ordinary differential equations, formed by the monodomain equation and the Courtemanche-Ramirez-Nattel model. Our new model parametrization is then adopted to predict the formation and self-sustainment of localized reentries characterizing atrial fibrillation, by numerically simulating the onset of ectopic beats from the pulmonary veins. We investigate the paroxysmal and the persistent form of AF starting from electro-anatomical maps of two patients. The model's response to stimulation shows how substrate characteristics play a key role in inducing and sustaining these arrhythmias. Localized reentries are less frequent and less stable in case of paroxysmal AF, while they tend to anchor themselves in areas affected by severe slow conduction in case of persistent AF.

show abstract

Graph Convolutional Regression of Cardiac Depolarization from Sparse Endocardial Maps

Meister

Passerini

Audigier

et al. 2021

Statistical Atlases and Computational Models of the Heart. M&Ms and EMIDEC Challenges

View full text Add to dashboard Cite

Estimation of Local Conduction Velocity from Myocardium Activation Time: Application to Cardiac Resynchronization Therapy

Cited by 6 publications

References 7 publications

Extrapolation of Ventricular Activation Times From Sparse Electroanatomical Data Using Graph Convolutional Neural Networks

Extrapolation of Ventricular Activation Times From Sparse Electroanatomical Data Using Graph Convolutional Neural Networks

A Computational Study of the Electrophysiological Substrate in Patients Suffering From Atrial Fibrillation

Graph Convolutional Regression of Cardiac Depolarization from Sparse Endocardial Maps

Contact Info

Product

Resources

About