Investigating a Novel Activation-Repolarisation Time Metric to Predict Localised Vulnerability to Reentry Using Computational Modelling

Enabling forward uncertainty quantification and sensitivity analysis in cardiac electrophysiology by reduced order modeling and machine learning

We present a new, computationally efficient framework to perform forward uncertainty quantification (UQ) in cardiac electrophysiology. We consider the monodomain model to describe the electrical activity in the cardiac tissue, coupled with the Aliev-Panfilov model to characterize the ionic activity through the cell membrane. We address a complete forward UQ pipeline, including both: (i) a variance-based global sensitivity analysis for the selection of the most relevant input parameters, and (ii) a way to perform uncertainty propagation to investigate the impact of intra-subject variability on outputs of interest depending on the cardiac potential. Both tasks exploit stochastic sampling techniques, thus implying overwhelming computational costs because of the huge amount of queries to the high-fidelity, full-order computational model obtained by approximating the coupled monodomain/Aliev-Panfilov system through the finite element method. To mitigate this computational burden, we replace the full-order model with computationally inexpensive projection-based reduced-order models (ROMs) aimed at reducing the state-space dimensionality. Resulting approximation errors on the outputs of interest are finally taken into account through artificial neural network (ANN)-based models, enhancing the accuracy of the whole UQ pipeline. Numerical results show that the proposed physics-based ROMs outperform regression-based emulators relying on ANNs built with the same amount of training data, in terms of both numerical accuracy and overall computational efficiency.

Assessing the ability of substrate mapping techniques to guide ventricular tachycardia ablation using computational modelling

BACKGROUND

Identification of targets for ablation of post-infarction ventricular tachycardias (VTs) remains challenging, often requiring arrhythmia induction to delineate the reentrant circuit. This carries a risk for the patient and may not be feasible. Substrate mapping has emerged as a safer strategy to uncover arrhythmogenic regions. However, VT recurrence remains common.

GOAL

To use computer simulations to assess the ability of different substrate mapping approaches to identify VT exit sites.

METHODS

A 3D computational model of the porcine post-infarction heart was constructed to simulate VT and paced rhythm. Electroanatomical maps were constructed based on endocardial electrogram features and the reentry vulnerability index (RVI - a metric combining activation (AT) and repolarization timings to identify tissue susceptibility to reentry). Since scar transmurality in our model was not homogeneous, parameters derived from all signals (including dense scar regions) were used in the analysis. Potential ablation targets obtained from each electroanatomical map during pacing were compared to the exit site detected during VT mapping.

RESULTS

Simulation data showed that voltage cut-offs applied to bipolar electrograms could delineate the scar, but not the VT circuit. Electrogram fractionation had the highest correlation with scar transmurality. The RVI identified regions closest to VT exit site but was outperformed by AT gradients combined with voltage cut-offs. The performance of all metrics was affected by pacing location.

CONCLUSIONS

Substrate mapping could provide information about the infarct, but the directional dependency on activation should be considered. Activation-repolarization metrics have utility in safely identifying VT targets, even with non-transmural scars.

The 'Digital Twin' to enable the vision of precision cardiology

Providing therapies tailored to each patient is the vision of precision medicine, enabled by the increasing ability to capture extensive data about individual patients. In this position paper, we argue that the second enabling pillar towards this vision is the increasing power of computers and algorithms to learn, reason, and build the 'digital twin' of a patient. Computational models are boosting the capacity to draw diagnosis and prognosis, and future treatments will be tailored not only to current health status and data, but also to an accurate projection of the pathways to restore health by model predictions. The early steps of the digital twin in the area of cardiovascular medicine are reviewed in this article, together with a discussion of the challenges and opportunities ahead. We emphasize the synergies between mechanistic and statistical models in accelerating cardiovascular research and enabling the vision of precision medicine.

In-silico study of the cardiac arrhythmogenic potential of biomaterial injection therapy

Biomaterial injection is a novel therapy to treat ischemic heart failure (HF) that has shown to reduce remodeling and restore cardiac function in recent preclinical studies. While the effect of biomaterial injection in reducing mechanical wall stress has been recently demonstrated, the influence of biomaterials on the electrical behavior of treated hearts has not been elucidated. In this work, we developed computational models of swine hearts to study the electrophysiological vulnerability associated with biomaterial injection therapy. The propagation of action potentials on realistic biventricular geometries was simulated by numerically solving the monodomain electrophysiology equations on anatomically-detailed models of normal, HF untreated, and HF treated hearts. Heart geometries were constructed from high-resolution magnetic resonance images (MRI) where the healthy, peri-infarcted, infarcted and gel regions were identified, and the orientation of cardiac fibers was informed from diffusion-tensor MRI. Regional restitution properties in each case were evaluated by constructing a probability density function of the action potential duration (APD) at different cycle lengths. A comparative analysis of the ventricular fibrillation (VF) dynamics for every heart was carried out by measuring the number of filaments formed after wave braking. Our results suggest that biomaterial injection therapy does not affect the regional dispersion of repolarization when comparing untreated and treated failing hearts. Further, we found that the treated failing heart is more prone to sustain VF than the normal heart, and is at least as susceptible to sustained VF as the untreated failing heart. Moreover, we show that the main features of VF dynamics in a treated failing heart are not affected by the level of electrical conductivity of the biogel injectates. This work represents a novel proof-of-concept study demonstrating the feasibility of computer simulations of the heart in understanding the arrhythmic behavior in novel therapies for HF.

Evaluation of the reentry vulnerability index to predict ventricular tachycardia circuits using high-density contact mapping

Background

Identifying arrhythmogenic sites to improve ventricular tachycardia (VT) ablation outcomes remains unresolved. The reentry vulnerability index (RVI) combines activation and repolarization timings to identify sites critical for reentrant arrhythmia initiation without inducing VT.

Objective

The purpose of this study was to provide the first assessment of RVI’s capability to identify VT sites of origin using high-density contact mapping and comparison with other activation-repolarization markers of functional substrate.

Methods

Eighteen VT ablation patients (16 male; 72% ischemic) were studied. Unipolar electrograms were recorded during ventricular pacing and analyzed offline. Activation time (AT), activation–recovery interval (ARI), and repolarization time (RT) were measured. Vulnerability to reentry was mapped based on RVI and spatial distribution of AT, ARI, and RT. The distance from sites identified as vulnerable to reentry to the VT site of origin was measured, with distances <10 mm and >20 mm indicating accurate and inaccurate localization, respectively.

Results

The origins of 18 VTs (6 entrainment, 12 pace-mapping) were identified. RVI maps included 1012 (408–2098) (median, 1st–3rd quartiles) points per patient. RVI accurately localized 72.2% VT sites of origin, with median distance of 5.1 (3.2–10.1) mm. Inaccurate localization was significantly less frequent for RVI than AT (5.6% vs 33.3%; odds ratio 0.12; P = .035). Compared to RVI, distance to VT sites of origin was significantly larger for sites showing prolonged RT and ARI and were nonsignificantly larger for sites showing highest AT and ARI gradients.

Conclusion

RVI identifies vulnerable regions closest to VT sites of origin. Activation-repolarization metrics may improve VT substrate delineation and inform novel ablation strategies.

Factors Promoting Conduction Slowing as Substrates for Block and Reentry in Infarcted Hearts

The development of effective and safe therapies for scar-related ventricular tachycardias requires a detailed understanding of the mechanisms underlying the conduction block that initiates electrical re-entries associated with these arrhythmias. Conduction block has been often associated with electrophysiological changes that prolong action potential duration (APD) within the border zone (BZ) of chronically infarcted hearts. However, experimental evidence suggests that remodeling processes promoting conduction slowing as opposed to APD prolongation mark the chronic phase. In this context, the substrate for the initial block at the mouth of an isthmus/diastolic channel leading to ventricular tachycardia is unclear. The goal of this study was to determine whether electrophysiological parameters associated with conduction slowing can cause block and re-entry in the BZ. In silico experiments were conducted on two-dimensional idealized infarct tissue as well as on a cohort of postinfarction porcine left ventricular models constructed from ex vivo magnetic resonance imaging scans. Functional conduction slowing in the BZ was modeled by reducing sodium current density, whereas structural conduction slowing was represented by decreasing tissue conductivity and including fibrosis. The arrhythmogenic potential of APD prolongation was also tested as a basis for comparison. Within all models, the combination of reduced sodium current with structural remodeling more often degenerated into re-entry and, if so, was more likely to be sustained for more cycles. Although re-entries were also detected in experiments with prolonged APD, they were often not sustained because of the subsequent block caused by long-lasting repolarization. Functional and structural conditions associated with slow conduction rather than APD prolongation form a potent substrate for arrhythmogenesis at the isthmus/BZ of chronically infarcted hearts. Reduced excitability led to block while slow conduction shortened the wavelength of propagation, facilitating the sustenance of re-entries. These findings provide important insights for models of patient-specific risk stratification and therapy planning.

Personalized Cardiac Computational Models: From Clinical Data to Simulation of Infarct-Related Ventricular Tachycardia

In the chronic stage of myocardial infarction, a significant number of patients develop life-threatening ventricular tachycardias (VT) due to the arrhythmogenic nature of the remodeled myocardium. Radiofrequency ablation (RFA) is a common procedure to isolate reentry pathways across the infarct scar that are responsible for VT. Unfortunately, this strategy show relatively low success rates; up to 50% of patients experience recurrent VT after the procedure. In the last decade, intensive research in the field of computational cardiac electrophysiology (EP) has demonstrated the ability of three-dimensional (3D) cardiac computational models to perform in-silico EP studies. However, the personalization and modeling of certain key components remain challenging, particularly in the case of the infarct border zone (BZ). In this study, we used a clinical dataset from a patient with a history of infarct-related VT to build an image-based 3D ventricular model aimed at computational simulation of cardiac EP, including detailed patient-specific cardiac anatomy and infarct scar geometry. We modeled the BZ in eight different ways by combining the presence or absence of electrical remodeling with four different levels of image-based patchy fibrosis (0, 10, 20, and 30%). A 3D torso model was also constructed to compute the ECG. Patient-specific sinus activation patterns were simulated and validated against the patient's ECG. Subsequently, the pacing protocol used to induce reentrant VTs in the EP laboratory was reproduced in-silico. The clinical VT was induced with different versions of the model and from different pacing points, thus identifying the slow conducting channel responsible for such VT. Finally, the real patient's ECG recorded during VT episodes was used to validate our simulation results and to assess different strategies to model the BZ. Our study showed that reduced conduction velocities and heterogeneity in action potential duration in the BZ are the main factors in promoting reentrant activity. Either electrical remodeling or fibrosis in a degree of at least 30% in the BZ were required to initiate VT. Moreover, this proof-of-concept study confirms the feasibility of developing 3D computational models for cardiac EP able to reproduce cardiac activation in sinus rhythm and during VT, using exclusively non-invasive clinical data.

Characterizing the clinical implementation of a novel activation-repolarization metric to identify targets for catheter ablation of ventricular tachycardias using computational models

Identification of targets for catheter ablation of ventricular tachycardias (VTs) remains a significant challenge. VTs are often driven by re-entrant circuits resulting from a complex interaction between the front (activation) and tail (repolarization) of the electrical wavefront. Most mapping techniques do not take into account the tissue repolarization which may hinder the detection of ablation targets. The re-entry vulnerability index (RVI), a recently proposed mapping procedure, incorporates both activation and repolarization times to uncover VT circuits. The method showed potential in a series of experiments, but it still requires further development to enable its incorporation into a clinical protocol. Here, in-silico experiments were conducted to thoroughly assess RVI maps constructed under clinically-relevant mapping conditions. Within idealized as well as anatomically realistic infarct models, we show that parameters of the algorithm such as the search radius can significantly alter the specificity and sensitivity of the RVI maps. When constructed on sparse grids obtained following various placements of clinical recording catheters, RVI maps can identify vulnerable regions as long as two electrodes were placed on both sides of the line of block. Moreover, maps computed during pacing without inducing VT can reveal areas of abnormal repolarization and slow conduction but not directly vulnerability. In conclusion, the RVI algorithm can detect re-entrant circuits during VT from low resolution mapping grids resembling the clinical setting. Furthermore, RVI maps may provide information about the underlying tissue electrophysiology to guide catheter ablation without the need of inducing potentially harmful VT during the clinical procedure. Finally, the ability of the RVI maps to identify vulnerable regions with specificity in high resolution computer models could potentially improve the prediction of optimal ablation targets of simulation-based strategies.

•

Safe and accurate detection of targets for catheter ablation remains a challenge.

•

We conducted a thorough assessment of the Re-entry Vulnerability Index (RVI).

•

Parameters of the algorithm can alter the specificity and sensitivity of RVI maps.

•

When constructed on sparse grids RVI maps could still detect arrhythmogenic sites.

•

In absence of arrhythmia, RVI maps revealed abnormal sites, but not vulnerability.

Patient-Specific Identification of Atrial Flutter Vulnerability-A Computational Approach to Reveal Latent Reentry Pathways

Atypical atrial flutter (AFlut) is a reentrant arrhythmia which patients frequently develop after ablation for atrial fibrillation (AF). Indeed, substrate modifications during AF ablation can increase the likelihood to develop AFlut and it is clinically not feasible to reliably and sensitively test if a patient is vulnerable to AFlut. Here, we present a novel method based on personalized computational models to identify pathways along which AFlut can be sustained in an individual patient. We build a personalized model of atrial excitation propagation considering the anatomy as well as the spatial distribution of anisotropic conduction velocity and repolarization characteristics based on a combination of a priori knowledge on the population level and information derived from measurements performed in the individual patient. The fast marching scheme is employed to compute activation times for stimuli from all parts of the atria. Potential flutter pathways are then identified by tracing loops from wave front collision sites and constricting them using a geometric snake approach under consideration of the heterogeneous wavelength condition. In this way, all pathways along which AFlut can be sustained are identified. Flutter pathways can be instantiated by using an eikonal-diffusion phase extrapolation approach and a dynamic multifront fast marching simulation. In these dynamic simulations, the initial pattern eventually turns into the one driven by the dominant pathway, which is the only pathway that can be observed clinically. We assessed the sensitivity of the flutter pathway maps with respect to conduction velocity and its anisotropy. Moreover, we demonstrate the application of tailored models considering disease-specific repolarization properties (healthy, AF-remodeled, potassium channel mutations) as well as applicabiltiy on a clinical dataset. Finally, we tested how AFlut vulnerability of these substrates is modulated by exemplary antiarrhythmic drugs (amiodarone, dronedarone). Our novel method allows to assess the vulnerability of an individual patient to develop AFlut based on the personal anatomical, electrophysiological, and pharmacological characteristics. In contrast to clinical electrophysiological studies, our computational approach provides the means to identify all possible AFlut pathways and not just the currently dominant one. This allows to consider all relevant AFlut pathways when tailoring clinical ablation therapy in order to reduce the development and recurrence of AFlut.

Assessment of a conduction-repolarisation metric to predict Arrhythmogenesis in right ventricular disorders

BACKGROUND

The re-entry vulnerability index (RVI) is a recently proposed activation-repolarization metric designed to quantify tissue susceptibility to re-entry. This study aimed to test feasibility of an RVI-based algorithm to predict the earliest endocardial activation site of ventricular tachycardia (VT) during electrophysiological studies and occurrence of haemodynamically significant ventricular arrhythmias in follow-up.

METHODS

Patients with Arrhythmogenic Right Ventricular Cardiomyopathy (ARVC) (n = 11), Brugada Syndrome (BrS) (n = 13) and focal RV outflow tract VT (n = 9) underwent programmed stimulation with unipolar electrograms recorded from a non-contact array in the RV.

RESULTS

Lowest values of RVI co-localised with VT earliest activation site in ARVC/BrS but not in focal VT. The distance between region of lowest RVI and site of VT earliest site (Dmin) was lower in ARVC/BrS than in focal VT (6.8 ± 6.7 mm vs 26.9 ± 13.3 mm, p = 0.005). ARVC/BrS patients with inducible VT had lower Global-RVI (RVIG) than those who were non-inducible (-54.9 ± 13.0 ms vs -35.9 ± 8.6 ms, p = 0.005) or those with focal VT (-30.6 ± 11.5 ms, p = 0.001). Patients were followed up for 112 ± 19 months. Those with clinical VT events had lower Global-RVI than both ARVC and BrS patients without VT (-54.5 ± 13.5 ms vs -36.2 ± 8.8 ms, p = 0.007) and focal VT patients (-30.6 ± 11.5 ms, p = 0.002).

CONCLUSIONS

RVI reliably identifies the earliest RV endocardial activation site of VT in BrS and ARVC but not focal ventricular arrhythmias and predicts the incidence of haemodynamically significant arrhythmias. Therefore, RVI may be of value in predicting VT exit sites and hence targeting of re-entrant arrhythmias.

Optimal contrast-enhanced MRI image thresholding for accurate prediction of ventricular tachycardia using ex-vivo high resolution models

Patient specific models created from contrast-enhanced (i.e. late-gadolinium, LGE) MRI images can be used for prediction of reentry location and clinical ablation planning. However, there is still a need for direct and systematic comparison between characteristics of ventricular tachycardia (VT) morphologies predicted in computational models and those acquired in clinical or experimental protocols. In this study, we aimed to: 1) assess the differences in VT morphologies predicted by modeling and recorded in experiments in terms of patterns and location of reentries, earliest and latest activation sites, and cycle lengths; and 2) define the optimal range of infarct tissue threshold values which provide best match between simulation and experimental results. To achieve these goals, we utilized LGE-MRI images from 4 swine hearts with inducible monomorphic VT. The images were segmented to identify non-infarcted myocardium, semi viable gray zone (GZ), and core scar based on pixel intensity. Several models were reconstructed from each LGE-MRI scan, with voxels of intensity between that of non-infarcted myocardium and 20-50% of the maximum intensity (in 10% increments) in the infarct region classified as GZ. VT induction was simulated in each model. Our simulation results showed that using GZ intensity thresholds of 20% or 30% resulted in the best match of simulated propagation patterns and reentry locations with those from the experiment. Overall, we matched 70% (7/10) morphologies for all the hearts. Our simulation shows that MRI-based computational models of hearts with myocardial infarction can accurately reproduce the majority of experimentally recorded post-infarction VTs.

Computational Representations of Myocardial Infarct Scars and Implications for Arrhythmogenesis

Image-based computational modeling is becoming an increasingly used clinical tool to provide insight into the mechanisms of reentrant arrhythmias. In the context of ischemic heart disease, faithful representation of the electrophysiological properties of the infarct region within models is essential, due to the scars known for arrhythmic properties. Here, we review the different computational representations of the infarcted region, summarizing the experimental measurements upon which they are based. We then focus on the two most common representations of the scar core (complete insulator or electrically passive tissue) and perform simulations of electrical propagation around idealized infarct geometries. Our simulations highlight significant differences in action potential duration and focal effective refractory period (ERP) around the scar, driven by differences in electrotonic loading, depending on the choice of scar representation. Finally, a novel mechanism for arrhythmia induction, following a focal ectopic beat, is demonstrated, which relies on localized gradients in ERP directly caused by the electrotonic sink effects of the neighboring passive scar.

Computational rabbit models to investigate the initiation, perpetuation, and termination of ventricular arrhythmia

Current understanding of cardiac electrophysiology has been greatly aided by computational work performed using rabbit ventricular models. This article reviews the contributions of multiscale models of rabbit ventricles in understanding cardiac arrhythmia mechanisms. This review will provide an overview of multiscale modeling of the rabbit ventricles. It will then highlight works that provide insights into the role of the conduction system, complex geometric structures, and heterogeneous cellular electrophysiology in diseased and healthy rabbit hearts to the initiation and maintenance of ventricular arrhythmia. Finally, it will provide an overview on the contributions of rabbit ventricular modeling on understanding the mechanisms underlying shock-induced defibrillation.