Hybrid Neural State-Space Modeling for Supervised and Unsupervised Electrocardiographic Imaging.

State-space modeling (SSM) provides a general framework for many image reconstruction tasks. Error in a priori physiological knowledge of the imaging physics, can bring incorrectness to solutions. Modern deep-learning approaches show great promise but lack interpretability and rely on large amounts of labeled data. In this paper, we present a novel hybrid SSM framework for electrocardiographic imaging (ECGI) to leverage the advantage of state-space formulations in data-driven learning. We first leverage the physics-based forward operator to supervise the learning. We then introduce neural modeling of the transition function and the associated Bayesian filtering strategy. We applied the hybrid SSM framework to reconstruct electrical activity on the heart surface from body-surface potentials. In unsupervised settings of both in-silico and in-vivo data without cardiac electrical activity as the ground truth to supervise the learning, we demonstrated improved ECGI performances of the hybrid SSM framework trained from a small number of ECG observations in comparison to the fixed SSM. We further demonstrated that, when in-silico simulation data becomes available, mixed supervised and unsupervised training of the hybrid SSM achieved a further 40.6% and 45.6% improvements, respectively, in comparison to traditional ECGI baselines and supervised data-driven ECGI baselines for localizing the origin of ventricular activations in real data.

Learning to Disentangle Inter-Subject Anatomical Variations in Electrocardiographic Data.

OBJECTIVE: This work investigates the possibility of disentangled representation learning of inter-subject anatomical variations within electrocardiographic (ECG) data. METHODS: Since ground truth anatomical factors are generally not known in clinical ECG for assessing the disentangling ability of the models, the presented work first proposes the SimECG data set, a 12-lead ECG data set procedurally generated with a controlled set of anatomical generative factors. Second, to perform such disentanglement, the presented method evaluates and compares deep generative models with latent density modeled by nonparametric Indian Buffet Process to account for the complex generative process of ECG data. RESULTS: In the simulated data, the experiments demonstrate, for the first time, concrete evidence of the possibility to disentangle key generative anatomical factors within ECG data in separation from task-relevant generative factors. We achieve a disentanglement score of 92.1% while disentangling five anatomical generative factors and the task-relevant generative factor. In both simulated and real-data experiments, this work further provides quantitative evidence for the benefit of disentanglement learning on the downstream clinical task of localizing the origin of ventricular activation. Overall, the presented method achieves an improvement of around 18.5%, and 11.3% for the simulated dataset, and around 7.2%, and 3.6% for the real dataset, over baseline CNN, and standard generative model, respectively. CONCLUSION: These results demonstrate the importance as well as the feasibility of the disentangled representation learning of inter-subject anatomical variations within ECG data. SIGNIFICANCE: This work suggests the important research direction to deal with the well-known challenge posed by the presence of significant inter-subject variations during an automated analysis of ECG data.

Fast Posterior Estimation of Cardiac Electrophysiological Model Parameters via Bayesian Active Learning.

Probabilistic estimation of cardiac electrophysiological model parameters serves an important step toward model personalization and uncertain quantification. The expensive computation associated with these model simulations, however, makes direct Markov Chain Monte Carlo (MCMC) sampling of the posterior probability density function (pdf) of model parameters computationally intensive. Approximated posterior pdfs resulting from replacing the simulation model with a computationally efficient surrogate, on the other hand, have seen limited accuracy. In this study, we present a Bayesian active learning method to directly approximate the posterior pdf function of cardiac model parameters, in which we intelligently select training points to query the simulation model in order to learn the posterior pdf using a small number of samples. We integrate a generative model into Bayesian active learning to allow approximating posterior pdf of high-dimensional model parameters at the resolution of the cardiac mesh. We further introduce new acquisition functions to focus the selection of training points on better approximating the shape rather than the modes of the posterior pdf of interest. We evaluated the presented method in estimating tissue excitability in a 3D cardiac electrophysiological model in a range of synthetic and real-data experiments. We demonstrated its improved accuracy in approximating the posterior pdf compared to Bayesian active learning using regular acquisition functions, and substantially reduced computational cost in comparison to existing standard or accelerated MCMC sampling.

Assessment of an ECG-Based System for Localizing Ventricular Arrhythmias in Patients With Structural Heart Disease.

Background We have previously developed an intraprocedural automatic arrhythmia-origin localization (AAOL) system to identify idiopathic ventricular arrhythmia origins in real time using a 3-lead ECG. The objective was to assess the localization accuracy of ventricular tachycardia (VT) exit and premature ventricular contraction (PVC) origin sites in patients with structural heart disease using the AAOL system. Methods and Results In retrospective and prospective case series studies, a total of 42 patients who underwent VT/PVC ablation in the setting of structural heart disease were recruited at 2 different centers. The AAOL system combines 120-ms QRS integrals of 3 leads (III, V2, V6) with pace mapping to predict VT exit/PVC origin site and projects that site onto the patient-specific electroanatomic mapping surface. VT exit/PVC origin sites were clinically identified by activation mapping and/or pace mapping. The localization error of the VT exit/PVC origin site was assessed by the distance between the clinically identified site and the estimated site. In the retrospective study of 19 patients with structural heart disease, the AAOL system achieved a mean localization accuracy of 6.5±2.6 mm for 25 induced VTs. In the prospective study with 23 patients, mean localization accuracy was 5.9±2.6 mm for 26 VT exit and PVC origin sites. There was no difference in mean localization error in epicardial sites compared with endocardial sites using the AAOL system (6.0 versus 5.8 mm, P=0.895). Conclusions The AAOL system achieved accurate localization of VT exit/PVC origin sites in patients with structural heart disease; its performance is superior to current systems, and thus, it promises to have potential clinical utility.

Prospective Multicenter Assessment of a New Intraprocedural Automated System for Localizing Idiopathic Ventricular Arrhythmia Origins.

OBJECTIVES: The objective of this study was to present a new system, the Automatic Arrhythmia Origin Localization (AAOL) system, which used incomplete electroanatomic mapping (EAM) for localization of idiopathic ventricular arrhythmia (IVA) origin on the patient-specific geometry of left ventricular, right ventricular, and neighboring vessels. The study assessed the accuracy of the system in localizing IVA source sites on cardiac structures where pace mapping is challenging. BACKGROUND: An intraprocedural automated site of origin localization system was previously developed to identify the origin of early left ventricular activation by using 12-lead electrocardiograms (ECGs). However, it has limitations, as it could not identify the site of origin in the right ventricle and relied on acquiring a complete EAM. METHODS: Twenty patients undergoing IVA catheter ablation had a 12-lead ECG recorded during clinical arrhythmia and during pacing at various locations identified on EAM geometries. The new system combined 3-lead (III, V2, and V6) 120-ms QRS integrals and patient-specific EAM geometry with pace mapping to predict the site of earliest ventricular activation. The predicted site was projected onto EAM geometry. RESULTS: Twenty-three IVA origin sites were clinically identified by activation mapping and/or pace mapping (8, right ventricle; 15, left ventricle, including 8 from the posteromedial papillary muscle, 2 from the aortic root, and 1 from the distal coronary sinus). The new system achieved a mean localization accuracy of 3.6 mm for the 23 mapped IVAs. CONCLUSIONS: The new intraprocedural AAOL system achieved accurate localization of IVA origin in ventricles and neighboring vessels, which could facilitate ablation procedures for patients with IVAs.

Feasibility study shows concordance between image-based virtual-heart ablation targets and predicted ECG-based arrhythmia exit-sites.

INTRODUCTION: We recently developed two noninvasive methodologies to help guide VT ablation: population-derived automated VT exit localization (PAVEL) and virtual-heart arrhythmia ablation targeting (VAAT). We hypothesized that while very different in their nature, limitations, and type of ablation targets (substrate-based vs. clinical VT), the image-based VAAT and the ECG-based PAVEL technologies would be spatially concordant in their predictions. OBJECTIVE: The objective is to test this hypothesis in ischemic cardiomyopathy patients in a retrospective feasibility study. METHODS: Four post-infarct patients who underwent LV VT ablation and had pre-procedural LGE-CMRs were enrolled. Virtual hearts with patient-specific scar and border zone identified potential VTs and ablation targets. Patient-specific PAVEL based on a population-derived statistical method localized VT exit sites onto a patient-specific 238-triangle LV endocardial surface. RESULTS: Ten induced VTs were analyzed and 9-exit sites were localized by PAVEL onto the patient-specific LV endocardial surface. All nine predicted VT exit sites were in the scar border zone defined by voltage mapping and spatially correlated with successful clinical lesions. There were 2.3 ± 1.9 VTs per patient in the models. All five VAAT lesions fell within regions ablated clinically. VAAT targets correlated well with 6 PAVEL-predicted VT exit sites. The distance between the center of the predicted VT-exit-site triangle and nearest corresponding VAAT ablation lesion was 10.7 ± 7.3 mm. CONCLUSIONS: VAAT targets are concordant with the patient-specific PAVEL-predicted VT exit sites. These findings support investigation into combining these two complementary technologies as a noninvasive, clinical tool for targeting clinically induced VTs and regions likely to harbor potential VTs.

Prospective Assessment of an Automated Intraprocedural 12-Lead ECG-Based System for Localization of Early Left Ventricular Activation.

BACKGROUND: To facilitate ablation of ventricular tachycardia (VT), an automated localization system to identify the site of origin of left ventricular activation in real time using the 12-lead ECG was developed. The objective of this study was to prospectively assess its accuracy. METHODS: The automated site of origin localization system consists of 3 steps: (1) localization of ventricular segment based on population templates, (2) population-based localization within a segment, and (3) patient-specific site localization. Localization error was assessed by the distance between the known reference site and the estimated site. RESULTS: In 19 patients undergoing 21 catheter ablation procedures of scar-related VT, site of origin localization accuracy was estimated using 552 left ventricular endocardial pacing sites pooled together and 25 VT-exit sites identified by contact mapping. For the 25 VT-exit sites, localization error of the population-based localization steps was within 10 mm. Patient-specific site localization achieved accuracy of within 3.5 mm after including up to 11 pacing (training) sites. Using 3 remotes (67.8±17.0 mm from the reference VT-exit site), and then 5 close pacing sites, resulted in localization error of 7.2±4.1 mm for the 25 identified VT-exit sites. In 2 emulated clinical procedure with 2 induced VTs, the site of origin localization system achieved accuracy within 4 mm. CONCLUSIONS: In this prospective validation study, the automated localization system achieved estimated accuracy within 10 mm and could thus provide clinical utility.

Embedding high-dimensional Bayesian optimization via generative modeling: Parameter personalization of cardiac electrophysiological models.

The estimation of patient-specific tissue properties in the form of model parameters is important for personalized physiological models. Because tissue properties are spatially varying across the underlying geometrical model, it presents a significant challenge of high-dimensional (HD) optimization at the presence of limited measurement data. A common solution to reduce the dimension of the parameter space is to explicitly partition the geometrical mesh. In this paper, we present a novel concept that uses a generative variational auto-encoder (VAE) to embed HD Bayesian optimization into a low-dimensional (LD) latent space that represents the generative code of HD parameters. We further utilize VAE-encoded knowledge about the generative code to guide the exploration of the search space. The presented method is applied to estimating tissue excitability in a cardiac electrophysiological model in a range of synthetic and real-data experiments, through which we demonstrate its improved accuracy and substantially reduced computational cost in comparison to existing methods that rely on geometry-based reduction of the HD parameter space.

Sequential Factorized Autoencoder for Localizing the Origin of Ventricular Activation From 12-Lead Electrocardiograms.

OBJECTIVE: This work presents a novel approach to handle the inter-subject variations existing in the population analysis of ECG, applied for localizing the origin of ventricular tachycardia (VT) from 12-lead electrocardiograms (ECGs). METHODS: The presented method involves a factor disentangling sequential autoencoder (f-SAE) - realized in both long short-term memory (LSTM) and gated recurrent unit (GRU) networks - to learn to disentangle the inter-subject variations from the factor relating to the location of origin of VT. To perform such disentanglement, a pair-wise contrastive loss is introduced. RESULTS: The presented methods are evaluated on ECG dataset with 1012 distinct pacing sites collected from scar-related VT patients during routine pace-mapping procedures. Experiments demonstrate that, for classifying the origin of VT into the predefined segments, the presented f-SAE improves the classification accuracy by 8.94% from using prescribed QRS features, by 1.5% from the supervised deep CNN network, and 5.15% from the standard SAE without factor disentanglement. Similarly, when predicting the coordinates of the VT origin, the presented f-SAE improves the performance by 2.25 mm from using prescribed QRS features, by 1.18 mm from the supervised deep CNN network and 1.6 mm from the standard SAE. CONCLUSION: These results demonstrate the importance as well as the feasibility of the presented f-SAE approach for separating inter-subject variations when using 12-lead ECG to localize the origin of VT. SIGNIFICANCE: This work suggests the important research direction to deal with the well-known challenge posed by inter-subject variations during population analysis from ECG signals.

Automated intraprocedural localization of origin of ventricular activation using patient-specific computed tomographic imaging.

BACKGROUND: To facilitate catheter ablation of ventricular tachycardia (VT), we previously developed an automated method to identify sources of left ventricular (LV) activation in real time using 12-lead electrocardiography (ECG), the accuracy of which depends on acquisition of a complete electroanatomic (EA) map. OBJECTIVE: The purpose of this study was to assess the feasibility of using a registered cardiac computed tomogram (CT) rather than an EA map to permit real-time localization and avoid errors introduced by incomplete maps. METHODS: Before LV VT ablation, 10 patients underwent CT imaging and 3-dimensional reconstruction of the cardiac surface to create a triangle mesh surface, which was registered to the EA map during the procedure and imported into custom localization software. The software uses QRS integrals from leads III, V2, and V6; derives personalized regression coefficients from pacing at ≥5 sites with known locations; and estimates the location of unknown activation sites on the 3-dimensional patient-specific LV endocardial surface. Localization accuracy was quantified for VT exit sites in millimeters by comparing the calculated against the known locations. RESULTS: The VT exit site was identified for 20 VTs using activation and entrainment mapping, supplemented by pace-mapping at the scar margin. The automated localization software achieved incremental accuracy with additional pacing sites and had a mean localization error of 6.9 ± 5.7 mm for the 20 VTs. CONCLUSION: Patient-specific CT geometry is feasible for use in real-time automated localization of ventricular activation and may avoid reliance on a complete EA map.

Noninvasive Reconstruction of Transmural Transmembrane Potential With Simultaneous Estimation of Prior Model Error.

To reconstruct electrical activity in the heart from body-surface electrocardiograms (ECGs) is an ill-posed inverse problem. Electrophysiological models have been found effective in regularizing these inverse problems by incorporating a priori knowledge about how the electrical potential in the heart propagates over time. However, these models suffer from model errors arising from, for example, parameters associated with tissue properties and the earliest sites of excitation. We present a Bayesian approach to simultaneously estimate transmembrane potential (TMP) signals and prior model errors, exploiting sparsity of the error in the gradient domain in the form of a novel sparse prior based on variational lower bound of the generalized Gaussian distribution. In synthetic and real-data experiments, we demonstrate the improvement of accuracy in TMP reconstruction brought by simultaneous model error estimation. We further provide theoretical and empirical justifications for the change of performances in the presented method at the presence of different model errors.

Localization of Activation Origin on Patient-Specific Epicardial Surface by Empirical Bayesian Method.

OBJECTIVE: Ablation treatment of ventricular arrhythmias can be facilitated by pre-procedure planning aided by electrocardiographic inverse solution, which can help to localize the origin of arrhythmia. Our aim was to improve localization accuracy of the inverse solution by using a novel Bayesian approach. METHODS: The inverse problem of electrocardiography was solved by reconstructing epicardial potentials from 120 body-surface electrocardiograms and from patient-specific geometry of the heart and torso for four patients suffering from scar-related ventricular tachycardia who underwent epicardial catheter mapping, which included pace-mapping. Simulations using dipole sources in patient-specific geometry were also performed. The proposed method, using dynamic spatio-temporal a priori constraints of the solution, was compared with classical Tikhonov methods based on fixed constraints. RESULTS: The mean localization error of the proposed method for all available pacing sites (n=78) was significantly smaller than that achieved by Tikhonov methods; specifically, the localization accuracy for pacing in the normal tissue (n=17) was [Formula: see text] mm (mean ± SD) versus [Formula: see text] mm reported in the previous study using the same clinical data and Tikhonov regularization. Simulation experiments further supported these clinical findings. CONCLUSION: The promising results of in vivo and in silico experiments presented in this study provide a strong incentive to pursuing further investigation of data-driven Bayesian methods in solving the electrocardiographic inverse problem. SIGNIFICANCE: The proposed approach to localizing origin of ventricular activation sequence may have important applications in pre-procedure assessment of arrhythmias and in guiding their ablation treatment.

Localization of Ventricular Activation Origin from the 12-Lead ECG: A Comparison of Linear Regression with Non-Linear Methods of Machine Learning.

We have previously developed an automated localization method based on multiple linear regression (MLR) model to estimate the activation origin on a generic left-ventricular (LV) endocardial surface in real time from the 12-lead ECG. The present study sought to investigate whether machine learning-namely, random-forest regression (RFR) and support-vector regression (SVR)-can improve the localization accuracy compared to MLR. For 38 patients the 12-lead ECG was acquired during LV endocardial pacing at 1012 sites with known coordinates exported from an electroanatomic mapping system; each pacing site was then registered to a generic LV endocardial surface subdivided into 16 segments tessellated into 238 triangles. ECGs were reduced to one variable per lead, consisting of 120-ms time integral of the QRS. To compare three regression models, the entire dataset ([Formula: see text]) was partitioned at random into a design set with 80% and a test set with the remaining 20% of the entire set, and the localization error-measured as geodesic distance on the generic LV surface-was assessed. Bootstrap method with replacement, using 1000 resampling trials, estimated each model's error distribution for the left-out sample ([Formula: see text]). In the design set ([Formula: see text]), the mean accuracy was 8.8, 12.1, and 12.9 mm, respectively for SVR, RVR and MLR. In the test set ([Formula: see text]), the mean value of the localization error in the SVR model was consistently lower than the other two models, both in comparison with the MLR (11.4 vs. 12.5 mm), and with the RFR (11.4 vs. 12.0 mm); the RFR model was also better than the MLR model for estimating localization accuracy (12.0 vs. 12.5 mm). The bootstrap method with 1,000 trials confirmed that the SVR and RFR models had significantly higher predictive accurate than the MLR in the bootstrap assessment with the left-out sample (SVR vs. MLR ([Formula: see text]), RFR vs. MLR ([Formula: see text])). The performance comparison of regression models showed that a modest improvement in localization accuracy can be achieved by SVR and RFR models, in comparison with MLR. The "population" coefficients generated by the optimized SVR model from our dataset are superior to the previously-derived "population" coefficients generated by the MLR model and can supersede them to improve the localization of ventricular activation on the generic LV endocardial surface.

Localization of Activation Origin on Patient-Specific Endocardial Surface by the Equivalent Double Layer (EDL) Source Model With Sparse Bayesian Learning.

OBJECTIVE: Ablation treatment of ventricular arrhythmias can be facilitated by pre-procedure planning aided by electrocardiographic inverse solution, which can help to localize the origin of arrhythmia. Our aim was to improve localization accuracy of the inverse solution for activation originating on the left-ventricular endocardial surface, by using a sparse Bayesian learning (SBL). METHODS: The inverse problem of electrocardiography was solved by reconstructing endocardial potentials from time integrals of body-surface electrocardiograms and from patient-specific geometry of the heart and torso for three patients with structurally normal ventricular myocardium, who underwent endocardial catheter mapping that included pace mapping. Complementary simulations using dipole sources in patient-specific geometry were also performed. The proposed method is using sparse property of the equivalent-double-layer (EDL) model of cardiac sources; it employs the SBL and makes use of the spatio-temporal features of the cardiac action potentials. RESULTS: The mean localization error of the proposed method for pooled pacing sites ( n=52) was significantly smaller ( p=0.0039) than that achieved for the same patients in the study of Erem et al. Simulation experiments localized the source dipoles ( n=48) from forward-simulated potentials with the error of 9.4 ± 4.5 mm (mean ± SD). CONCLUSION: The results of our clinical and simulation experiments demonstrate that localization of left-ventricular endocardial activation by means of the Bayesian approach, based on sparse representation of sources by EDL, is feasible and accurate. SIGNIFICANCE: The proposed approach to localizing endocardial sources may have important applications in pre-procedure assessment of arrhythmias and in guiding their ablation treatment.

Rapid 12-lead automated localization method: Comparison to electrocardiographic imaging (ECGI) in patient-specific geometry.

BACKGROUND: Rapid accurate localization of the site of ventricular activation origin during catheter ablation for ventricular arrhythmias could facilitate the procedure. Electrocardiographic imaging (ECGI) using large lead sets can localize the origin of ventricular activation. We have developed an automated method to identify sites of early ventricular activation in real time using the 12-lead ECG. We aim to compare the localization accuracy of ECGI and the automated method, identifying pacing sites/VT exit based on a patient-specific model. METHODS: A patient undergoing ablation of VT on the left-ventricular endocardium and epicardium had 120-lead body-surface potential mapping (BSPM) recorded during the procedure. (1) ECGI methodology: The L1-norm regularization was employed to reconstruct epicardial potentials based on patient-specific geometry for localizing endocardial ventricular activation origin. We used the BSPM data corresponding to known endocardial pacing sites and a VT exit site identified by 3D contact mapping to analyze them offline. (2) The automatedmethod: location coordinates of pacing sites together with the time integral of the first 120 ms of the QRS complex of 3 ECG predictors (leads III, V2 and V6) were used to calculate patient-specific regression coefficients to predict the location of unknown sites of ventricular activation origin ("target" sites). Localization error was quantified over all pacing sites in millimeters by comparing the calculated location and the known reference location. RESULTS: Localization was tested for 14 endocardial pacing sites and 1 epicardial VT exit site. For 14 endocardial pacing sites the mean localization error of the automated method was significantly lower than that of the ECGI (8.9 vs. 24.9 mm, p < 0.01), when 10 training pacing sites are used. Emulation of a clinical procedure demonstrated that the automated method achieved localization error of <5 mm for the VT-exit site; while the ECGI approach approximately correlates with the site of VT exit from the scar within a distance of 18.4 mm. CONCLUSIONS: The automated method using only 3 ECGs shows promise to localize the origin of ventricular activation as tested by pacing, and the VT-exit site and compares favourably to inverse solution calculation, avoiding cumbersome lead sets. As 12-lead ECG data is acquired by current 3D mapping systems, it is conceivable that the algorithm could be directly incorporated into a mapping system. Further validation in a prospective cohort study is needed to confirm and extend observations reported in this study.

Criteria for ECG detection of acute myocardial ischemia: Sensitivity versus specificity.

BACKGROUND: Criteria for electrocardiographic detection of acute myocardial ischemia recommended by the Consensus Document of ESC/ACCF/AHA/WHF consist of two parts: The ST elevation myocardial infarction (STEMI) criteria based on ST elevation (ST↑) in 10 pairs of contiguous leads and the other on ST depression (ST↓) in the same 10 contiguous pairs. Our aim was to assess sensitivity (SE) and specificity (SP) of these criteria-and to seek their possible improvements-in three databases of 12­lead ECGs. METHODS: We used (1) STAFF III data of controlled ischemic episodes recorded from 99 patients (pts) during percutaneous coronary intervention (PCI) involving either left anterior descending (LAD) coronary artery, right coronary artery (RCA), or left circumflex (LCx) coronary artery. (2) Data from the University of Glasgow for 58 pts with acute myocardial infarction (AMI) and 58 pts without AMI, as confirmed by MRI. (3) Data from Lund University retrieved from a centralized ECG management system for 100 pts with various pathological ST changes-other than acute coronary occlusion-including ventricular pre-excitation, acute pericarditis, early repolarization syndrome, left ventricular hypertrophy, and left bundle branch block. ST measurements at J-point in ECGs of all 315 pts were obtained automatically on the averaged beat with manual review and the recommended criteria as well as their proposed modifications, were applied. Performance measures included SE, SP, positive predictive value (PPV), and benefit-to-harm ratio (BHR), defined as the ratio of true-positive vs. false-positive detections. RESULTS: We found that the SE of widely-used STEMI criteria can be indeed improved by the additional ST↓ criteria, but at the cost of markedly decreased SP. In contrast, using ST↑ in only 3 additional contiguous pairs of leads (STEMI13) can boost SE without any loss of SP. In the STAFF III database, SE/SP/PPV were 56/98/97% for the STEMI, 79/79/79% for the STEMI with added ST↓ and 67/97/96% for the STEMI13. In the Glasgow database, corresponding SE/SP/PPV were 43/98/96%, 84/90/89%, and 55/98/97%. For the Lund database, SP was 56% for the STEMI, 24% for the STEMI with ST↓, and 56% for the STEMI13. CONCLUSION: Current recommended criteria for detecting acute myocardial ischemia, involving ST↓, boost SE of widely-used STEMI criteria, at the cost of SP. To keep the SP high, we propose either the adjustment of threshold for the added ST↓ criteria or a selective use of ST↓ only in contiguous leads V2 and V3 plus ST↑ in lead pairs (aVL, -III) and (III, -aVL).

Quantifying the uncertainty in model parameters using Gaussian process-based Markov chain Monte Carlo in cardiac electrophysiology.

Model personalization requires the estimation of patient-specific tissue properties in the form of model parameters from indirect and sparse measurement data. Moreover, a low-dimensional representation of the parameter space is needed, which often has a limited ability to reveal the underlying tissue heterogeneity. As a result, significant uncertainty can be associated with the estimated values of the model parameters which, if left unquantified, will lead to unknown variability in model outputs that will hinder their reliable clinical adoption. Probabilistic estimation of model parameters, however, remains an unresolved challenge. Direct Markov Chain Monte Carlo (MCMC) sampling of the posterior distribution function (pdf) of the parameters is infeasible because it involves repeated evaluations of the computationally expensive simulation model. To accelerate this inference, one popular approach is to construct a computationally efficient surrogate and sample from this approximation. However, by sampling from an approximation, efficiency is gained at the expense of sampling accuracy. In this paper, we address this issue by integrating surrogate modeling of the posterior pdf into accelerating the Metropolis-Hastings (MH) sampling of the exact posterior pdf. It is achieved by two main components: (1) construction of a Gaussian process (GP) surrogate of the exact posterior pdf by actively selecting training points that allow for a good global approximation accuracy with a focus on the regions of high posterior probability; and (2) use of the GP surrogate to improve the proposal distribution in MH sampling, in order to improve the acceptance rate. The presented framework is evaluated in its estimation of the local tissue excitability of a cardiac electrophysiological model in both synthetic data experiments and real data experiments. In addition, the obtained posterior distributions of model parameters are interpreted in relation to the factors contributing to parameter uncertainty, including different low-dimensional representations of the parameter space, parameter non-identifiability, and parameter correlations.

Localization of ventricular activation origin using patient-specific geometry: Preliminary results.

BACKGROUND AND OBJECTIVES: Catheter ablation of ventricular tachycardia (VT) may include induction of VT and localization of VT-exit site. Our aim was to assess localization performance of a novel statistical pace-mapping method and compare it with performance of an electrocardiographic inverse solution. METHODS: Seven patients undergoing ablation of VT (4 with epicardial, 3 with endocardial exit) aided by electroanatomic mapping underwent intraprocedural 120-lead body-surface potential mapping (BSPM). Two approaches to localization of activation origin were tested: (1) A statistical method, based on multiple linear regression (MLR), which required only the conventional 12-lead ECG for a sufficient number of pacing sites with known origin together with patient-specific geometry of the endocardial/epicardial surface obtained by electroanatomic mapping; and (2) a classical deterministic inverse solution for recovering heart-surface potentials, which required BSPM and patient-specific geometry of the heart and torso obtained via computed tomography (CT). RESULTS: For the MLR method, at least 10-15 pacing sites with known coordinates, together with their corresponding 12-lead ECGs, were required to derive reliable patient-specific regression equations, which then enabled accurate localization of ventricular activation with unknown origin. For 4 patients who underwent epicardial mapping, the median of localization error for the MLR was significantly lower than that for the inverse solution (10.6 vs. 27.3 mm, P  =  0.034); a similar result held for 3 patients who underwent endocardial mapping (7.7 vs. 17.1 mm, P  =  0.017). The pooled localization error for all epicardial and endocardial sites was also significantly smaller for the MLR compared with the inverse solution (P  =  0.005). CONCLUSIONS: The novel pace-mapping approach to localizing the origin of ventricular activation offers an easily implementable supplement and/or alternative to the preprocedure inverse solution; its simplicity makes it suitable for real-time applications during clinical catheter-ablation procedures.

Non-invasive epicardial and endocardial electrocardiographic imaging for scar-related ventricular tachycardia.

Aims: Contact mapping is currently used to guide catheter ablation of scar-related ventricular tachycardia (VT) but usually provides incomplete assessment of 3D re-entry circuits and their arrhythmogenic substrates. This study investigates the feasibility of non-invasive electrocardiographic imaging (ECGi) in mapping scar substrates and re-entry circuits throughout the epicardium and endocardium. Methods and results: Four patients undergoing endocardial and epicardial mapping and ablation of scar-related VT had computed tomography scans and a 120-lead electrocardiograms, which were used to compute patient-specific ventricular epicardial and endocardial unipolar electrograms (CEGMs). Native-rhythm CEGMs were used to identify sites of myocardial scar and signal fractionation. Computed electrograms of induced VT were used to localize re-entrant circuits and exit sites. Results were compared to in vivo contact mapping data and epicardium-based ECGi solutions. During native rhythm, an average of 493 ± 18 CEGMs were analysed on each patient. Identified regions of scar and fractionation comprised, respectively, 25 ± 4% and 2 ± 1% of the ventricular surface area. Using a linear mixed-effects model grouped at the level of an individual patient, CEGM voltage and duration were significantly associated with contact bipolar voltage. During induced VT, the inclusion of endocardial layer in ECGi made it possible to identify two epicardial vs. three endocardial VT exit sites among five reconstructed re-entry circuits. Conclusion: Electrocardiographic imaging may be used to reveal sites of signal fractionation and to map short-lived VT circuits. Its capacity to map throughout epicardial and endocardial layers may improve the delineation of 3D re-entry circuits and their arrhythmogenic substrates.

Novel Metric Using Laplacian Eigenmaps to Evaluate Ischemic Stress on the Torso Surface.

The underlying pathophysiology of myocardial ischemia is incompletely understood, resulting in persistent difficulty of diagnosis. This limited understanding of underlying mechanisms encourages a data driven approach, which seeks to identify patterns in the ECG data that can be linked statistically to disease states. Laplacian Eigen-maps (LE) is a dimensionality reduction method popularized in machine learning that we have shown in large animal experiments to identify underlying ischemic stress both earlier in an ischemic episode, and more robustly, than typical clinical markers. We have now extended this approach to body surface potential mapping (BSPM) recordings acquired during acute, transient ischemia episodes from animal and human PTCA studies. Our previous studies, suggest that the LE approach is sensitive to the spatiotemporal electrocardiographic consequences of ischemia-induced stress within the heart and on the epicardial surface. In this study, we expand this technique to the body surface of animals and humans. Across 10 episodes of induced ischemia in animals and 200 human recordings during PTCA, the LE algorithm was able to detect ischemic events from BSPM as changes in the morphology of the resulting trajectories while maintaining the superior temporal performance the LE-metric has shown previously.