Artificial intelligence for detection of ventricular oversensing: Machine learning approaches for noise detection within nonsustained ventricular tachycardia episodes remotely transmitted by pacemakers and implantable cardioverter-defibrillators.

BACKGROUND: Pacemakers (PMs) and implantable cardioverter-defibrillators (ICDs) increasingly automatically record and remotely transmit nonsustained ventricular tachycardia (NSVT) episodes, which may reveal ventricular oversensing. OBJECTIVES: We aimed to develop and validate a machine learning algorithm that accurately classifies NSVT episodes transmitted by PMs and ICDs in order to lighten health care workload burden and improve patient safety. METHODS: PMs or ICDs (Boston Scientific, St Paul, MN) from 4 French hospitals with ≥1 transmitted NSVT episode were split into 3 subgroups: training set, validation set, and test set. Each NSVT episode was labeled as either physiological or nonphysiological. Four machine learning algorithms-2DTF-CNN, 2D-DenseNet, 2DTF-VGG, and 1D-AgResNet-were developed using training and validation data sets. Accuracies of the classifiers were compared with an analysis of the remote monitoring team of the Bordeaux University Hospital using F2 scores (favoring sensitivity over predictive positive value) using an independent test set. RESULTS: A total of 807 devices transmitted 10,471 NSVT recordings (82% ICD; 18% PM), of which 87 devices (10.8%) transmitted 544 NSVT recordings with nonphysiological signals. The classification by the remote monitoring team resulted in an F2 score of 0.932 (sensitivity 95%; specificity 99%) The 4 machine learning algorithms showed high and comparable F2 scores (2DTF-CNN: 0.914; 2D-DenseNet: 0.906; 2DTF-VGG: 0.863; 1D-AgResNet: 0.791), and only 1D-AgResNet had significantly different labeling from that of the remote monitoring team. CONCLUSION: Machine learning algorithms were accurate in detecting nonphysiological signals within electrograms transmitted by PMs and ICDs. An artificial intelligence approach may render remote monitoring less resourceful and improve patient safety.

Effect of Segmentation Uncertainty on the ECGI Inverse Problem Solution and Source Localization.

Electrocardiographic Imaging (ECGI) is a promising tool to non-invasively map the electrical activity of the heart using body surface potentials (BSPs) and the patient specific anatomical data. One of the first steps of ECGI is the segmentation of the heart and torso geometries. In the clinical practice, the segmentation procedure is not fully-automated yet and is in consequence operator-dependent. We expect that the inter-operator variation in cardiac segmentation would influence the ECGI solution. This effect remains however non quantified. In the present work, we study the effect of segmentation variability on the ECGI estimation of the cardiac activity with 262 shape models generated from fifteen different segmentations. Therefore, we designed two test cases: with and without shape model uncertainty. Moreover, we used four cases for ectopic ventricular excitation and compared the ECGI results in terms of reconstructed activation times and excitation origins. The preliminary results indicate that a small variation of the activation maps can be observed with a model uncertainty but no significant effect on the source localization is observed.

The Effect of Segmentation Variability in Forward ECG Simulation.

Segmentation of patient-specific anatomical models is one of the first steps in Electrocardiographic imaging (ECGI). However, the effect of segmentation variability on ECGI remains unexplored. In this study, we assess the effect of heart segmentation variability on ECG simulation. We generated a statistical shape model from segmentations of the same patient and generated 262 cardiac geometries to run in an ECG forward computation of body surface potentials (BSPs) using an equivalent dipole layer cardiac source model and 5 ventricular stimulation protocols. Variability between simulated BSPs for all models and protocols was assessed using Pearson's correlation coefficient (CC). Compared to the BSPs of the mean cardiac shape model, the lowest variability (average CC = 0.98 ± 0.03) was found for apical pacing whereas the highest variability (average CC = 0.90 ± 0.23) was found for right ventricular free wall pacing. Furthermore, low amplitude BSPs show a larger variation in QRS morphology compared to high amplitude signals. The results indicate that the uncertainty in cardiac shape has a significant impact on ECGI.

Prediction of Clinical Deep Brain Stimulation Target for Essential Tremor From 1.5 Tesla MRI Anatomical Landmarks.

Background: Deep brain stimulation is an efficacious treatment for refractory essential tremor, though targeting the intra-thalamic nuclei remains challenging. Objectives: We sought to develop an inverse approach to retrieve the position of the leads in a cohort of patients operated on with optimal clinical outcomes from anatomical landmarks identifiable by 1.5 Tesla magnetic resonance imaging. Methods: The learning database included clinical outcomes and post-operative imaging from which the coordinates of the active contacts and those of anatomical landmarks were extracted. We used machine learning regression methods to build three different prediction models. External validation was performed according to a leave-one-out cross-validation. Results: Fifteen patients (29 leads) were included, with a median tremor improvement of 72% on the Fahn-Tolosa-Marin scale. Kernel ridge regression, deep neural networks, and support vector regression (SVR) were used. SVR gave the best results with a mean error of 1.33 ± 1.64 mm between the predicted target and the active contact position. Conclusion: We report an original method for the targeting in deep brain stimulation for essential tremor based on patients' radio-anatomical features. This approach will be tested in a prospective clinical trial.

Cardiac Activation Maps Reconstruction: A Comparative Study Between Data-Driven and Physics-Based Methods.

One of the essential diagnostic tools of cardiac arrhythmia is activation mapping. Noninvasive current mapping procedures include electrocardiographic imaging. It allows reconstructing heart surface potentials from measured body surface potentials. Then, activation maps are generated using the heart surface potentials. Recently, a study suggests to deploy artificial neural networks to estimate activation maps directly from body surface potential measurements. Here we carry out a comparative study between the data-driven approach DirectMap and noninvasive classic technique based on reconstructed heart surface potentials using both Finite element method combined with L1-norm regularization (FEM-L1) and the spatial adaptation of Time-delay neural networks (SATDNN-AT). In this work, we assess the performance of the three approaches using a synthetic single paced-rhythm dataset generated on the atria surface. The results show that data-driven approach DirectMap quantitatively outperforms the two other methods. In fact, we observe an absolute activation time error and a correlation coefficient, respectively, equal to 7.20 ms, 93.2% using DirectMap, 14.60 ms, 76.2% using FEM-L1 and 13.58 ms, 79.6% using SATDNN-AT. In addition, results show that data-driven approaches (DirectMap and SATDNN-AT) are strongly robust against additive gaussian noise compared to FEM-L1.

Uncertainty Quantification of the Effects of Segmentation Variability in ECGI.

Despite advances in many of the techniques used in Electrocardiographic Imaging (ECGI), uncertainty remains insufficiently quantified for many aspects of the pipeline. The effect of geometric uncertainty, particularly due to segmentation variability, may be the least explored to date. We use statistical shape modeling and uncertainty quantification (UQ) to compute the effect of segmentation variability on ECGI solutions. The shape model was made with Shapeworks from nine segmentations of the same patient and incorporated into an ECGI pipeline. We computed uncertainty of the pericardial potentials and local activation times (LATs) using polynomial chaos expansion (PCE) implemented in UncertainSCI. Uncertainty in pericardial potentials from segmentation variation mirrored areas of high variability in the shape model, near the base of the heart and the right ventricular outflow tract, and that ECGI was less sensitive to uncertainty in the posterior region of the heart. Subsequently LAT calculations could vary dramatically due to segmentation variability, with a standard deviation as high as 126ms, yet mainly in regions with low conduction velocity. Our shape modeling and UQ pipeline presented possible uncertainty in ECGI due to segmentation variability and can be used by researchers to reduce said uncertainty or mitigate its effects. The demonstrated use of statistical shape modeling and UQ can also be extended to other types of modeling pipelines.

A Cardiac Shape Model for Segmentation Uncertainty Quantification.

Segmentation of cardiac images is a variable component of many patient specific computational pipelines, yet its impact on simulated results are still not fully understood. A hurdle to to exploring the impact of the segmentation variability is the technical challenge of building a statistical shape model of the ventricles. In this study, we improved open our previous shape analysis by creating a unified shape model including both the epicardium and endocardium. We tested four techniques within ShapeWorks to generate a ventricular shape model: standard, multidomain, hybrid multidomain, and geodesic distance. The multidomain and hybrid multidomain generated a shape model using all eleven segmentations, and the geodesic distance method generated a shape model using a subset of four segmentations. Each of the shape models captured spatially dependent characteristics of the segmentation variability, including wall thickness, annular diameter, and basal truncation. While each of the three methods have benefits, the hybrid multidomain approach provided the most accurate shape model with fewest points and may be most useful in a majority of applications.

Virtual FFR Quantified with a Generalized Flow Model Using Windkessel Boundary Conditions.

Fractional flow reserve (FFR) has proved its efficiency in improving patient diagnosis. In this paper, we consider a 2D reconstructed left coronary tree with two artificial lesions of different degrees. We use a generalized fluid model with a Carreau law and use a coupled multidomain method to implement Windkessel boundary conditions at the outlets. We introduce our methodology to quantify the virtual FFR and conduct several numerical experiments. We compare FFR results from the Navier-Stokes model versus generalized flow model and for Windkessel versus traction-free outlet boundary conditions or mixed outlet boundary conditions. We also investigate some sources of uncertainty that the FFR index might encounter during the invasive procedure, in particular, the arbitrary position of the distal sensor. The computational FFR results show that the degree of stenosis is not enough to classify a lesion, while there is a good agreement between the Navier-Stokes model and the non-Newtonian flow model adopted in classifying coronary lesions. Furthermore, we highlight that the lack of standardization while making FFR measurement might be misleading regarding the significance of stenosis.

Simulating the effect of sodium channel blockage on cardiac electromechanics.

There is growing interest to better understand drug-induced cardiovascular complications and to predict undesirable side effects at as early a stage in the drug development process as possible. The purpose of this paper is to investigate computationally the influence of sodium ion channel blockage on cardiac electromechanics. To do so, we implement a myofiber orientation dependent passive stress model (Holzapfel-Ogden) in the multiphysics solver Chaste to simulate an imaged physiological model of the human ventricles. A dosage of a sodium channel blocker was then applied and its inhibitory effects on the electrical propagation across ventricles were modeled. We employ the Kerckhoffs active stress model to generate electrically excited contractile behavior of myofibers. Our predictions indicate that a delay in the electrical activation of ventricular tissue caused by the sodium channel blockage translates to a delay in the mechanical biomarkers that were investigated. Moreover, sodium channel blockage was found to increase left ventricular twist. A multiphysics computational framework from the cell level to the organ level was thus used to predict the effect of sodium channel blocking drugs on cardiac electromechanics.

Evaluation of Fifteen Algorithms for the Resolution of the Electrocardiography Imaging Inverse Problem Using ex-vivo and in-silico Data.

The electrocardiographic imaging inverse problem is ill-posed. Regularization has to be applied to stabilize the problem and solve for a realistic solution. Here, we assess different regularization methods for solving the inverse problem. In this study, we assess (i) zero order Tikhonov regularization (ZOT) in conjunction with the Method of Fundamental Solutions (MFS), (ii) ZOT regularization using the Finite Element Method (FEM), and (iii) the L1-Norm regularization of the current density on the heart surface combined with FEM. Moreover, we apply different approaches for computing the optimal regularization parameter, all based on the Generalized Singular Value Decomposition (GSVD). These methods include Generalized Cross Validation (GCV), Robust Generalized Cross Validation (RGCV), ADPC, U-Curve and Composite REsidual and Smoothing Operator (CRESO) methods. Both simulated and experimental data are used for this evaluation. Results show that the RGCV approach provides the best results to determine the optimal regularization parameter using both the FEM-ZOT and the FEM-L1-Norm. However for the MFS-ZOT, the GCV outperformed all the other regularization parameter choice methods in terms of relative error and correlation coefficient. Regarding the epicardial potential reconstruction, FEM-L1-Norm clearly outperforms the other methods using the simulated data but, using the experimental data, FEM based methods perform as well as MFS. Finally, the use of FEM-L1-Norm combined with RGCV provides robust results in the pacing site localization.

Effect of Segmentation Variation on ECG Imaging.

ECG imaging (ECGI) is the process of calculating electrical cardiac activity from body surface recordings from the geometry and conductivity of the torso volume. A key first step to create geometric models for ECGI and a possible source of considerable variability is to segment the surface of the heart. We hypothesize that this variation in cardiac segmentation will produce variation in the computed ventricular surface potentials from ECGI. To evaluate this hypothesis, we leveraged the resources of the Consortium for ECG Imaging (CEI) to carry out a comparison of ECGI results from the same body surface potentials and multiple ventricular segmentations. We found that using the different segmentations produced variability in the computed ventricular surface potentials. Not surprisingly, locations of greater variance in the computed potential correlated to locations of greater variance in the segmentations, for example near the pulmonary artery and basal anterior left ventricular wall. Our results indicate that ECGI may be more sensitive to segmentation errors on the anterior epicardial surface than on other areas of the heart.

Identification of Ion Currents Components Generating Field Potential Recorded in MEA From hiPSC-CM.

OBJECTIVE: Multi electrodes arrays (MEAs) combined with cardiomyocytes derived from human-induced pluripotent stem cells (hiPSC-CMs) can enable high- or medium-throughput drug screening in safety pharmacology. This technology has recently attracted a lot of attention, in particular from an international initiative named CiPA. But it is currently limited by the difficulty to analyze the measured signals. We propose a strategy to analyze the signals acquired by the MEA and to automatically deduce the channels affected by the drug. METHODS: Our method is based on the bidomain equations, a model for the MEA electrodes, and an inverse problem strategy. RESULTS: in silico MEA signals are obtained for two commercial devices and an example of early after depolarization is presented. Then, by processing real signals obtained for four different compounds, our algorithm was able to provide dose-response curves for potassium, sodium, and calcium channels. For ivabradine and moxifloxacin, the IC50 and dose-response curves are in very good agreement with known values. SIGNIFICANCE: The proposed strategy offers a possible answer to a major question raised by the community of safety pharmacology. By allowing a more automated analysis of the signals, our approach could contribute to promote the technology based on MEA and hiPSC-CMs and, therefore, improve reliability and efficiency of drug screening.

In silico assessment of the effects of various compounds in MEA/hiPSC-CM assays: Modeling and numerical simulations.

We propose a mathematical approach for the analysis of drugs effects on the electrical activity of human induced pluripotent stem cell-derived cardiomyocytes (hiPSC-CMs) based on multi-electrode array (MEA) experiments. Our goal is to produce an in silico tool able to simulate drugs action in MEA/hiPSC-CM assays. The mathematical model takes into account the geometry of the MEA and the electrodes' properties. The electrical activity of the stem cells at the ion-channel level is governed by a system of ordinary differential equations (ODEs). The ODEs are coupled to the bidomain equations, describing the propagation of the electrical wave in the stem cells preparation. The field potential (FP) measured by the MEA is modeled by the extracellular potential of the bidomain equations. First, we propose a strategy allowing us to generate a field potential in good agreement with the experimental data. We show that we are able to reproduce realistic field potentials by introducing different scenarios of heterogeneity in the action potential. This heterogeneity reflects the differentiation atria/ventricles and the age of the cells. Second, we introduce a drug/ion channels interaction based on a pore block model. We conduct different simulations for five drugs (mexiletine, dofetilide, bepridil, ivabradine and BayK). We compare the simulation results with the field potential collected from experimental measurements. Different biomarkers computed on the FP are considered, including depolarization amplitude, repolarization delay, repolarization amplitude and depolarization-repolarization segment. The simulation results show that the model reflect properly the main effects of these drugs on the FP.

A conduction velocity adapted eikonal model for electrophysiology problems with re-excitability evaluation.

Computational models of heart electrophysiology achieved a considerable interest in the medical community as they represent a novel framework for the study of the mechanisms underpinning heart pathologies. The high demand of computational resources and the long computational time required to evaluate the model solution hamper the use of detailed computational models in clinical applications. In this paper, we present a multi-front eikonal algorithm that adapts the conduction velocity (CV) to the activation frequency of the tissue substrate. We then couple the eikonal new algorithm with the Mitchell-Schaeffer (MS) ionic model to determine the tissue electrical state. Compared to the standard eikonal model, this model introduces three novelties: first, it evaluates the local value of the transmembrane potential and of the ionic variable solving an ionic model; second, it computes the action potential duration (APD) and the diastolic interval (DI) from the solution of the MS model and uses them to determine if the tissue is locally re-excitable; third, it adapts the CV to the underpinning electrophysiological state through an analytical expression of the CV restitution and the computed local DI. We conduct series of simulations on a 3D tissue slab and on a realistic heart geometry and compare the solutions with those obtained solving the monodomain equation. Our results show that the new model is significantly more accurate than the standard eikonal model. The proposed model enables the numerical simulation of the heart electrophysiology on a clinical time scale and thus constitutes a viable model candidate for computer-guided radio-frequency ablation.

Human ventricular activation sequence and the simulation of the electrocardiographic QRS complex and its variability in healthy and intraventricular block conditions.

AIMS: To investigate how variability in activation sequence and passive conduction properties translates into clinical variability in QRS biomarkers, and gain novel physiological knowledge on the information contained in the human QRS complex. METHODS AND RESULTS: Multiscale bidomain simulations using a detailed heart-torso human anatomical model are performed to investigate the impact of activation sequence characteristics on clinical QRS biomarkers. Activation sequences are built and validated against experimentally-derived ex vivo and in vivo human activation data. R-peak amplitude exhibits the largest variability in terms of QRS morphology, due to its simultaneous modulation by activation sequence speed, myocardial intracellular and extracellular conductivities, and propagation through the human torso. QRS width, however, is regulated by endocardial activation speed and intracellular myocardial conductivities, whereas QR intervals are only affected by the endocardial activation profile. Variability in the apico-basal location of activation sites on the anterior and posterior left ventricular wall is associated with S-wave progression in limb and precordial leads, respectively, and occasional notched QRS complexes in precordial derivations. Variability in the number of early activation sites successfully reproduces pathological abnormalities of the human conduction system in the QRS complex. CONCLUSION: Variability in activation sequence and passive conduction properties captures and explains a large part of the clinical variability observed in the human QRS complex. Our physiological insights allow for a deeper interpretation of human QRS biomarkers in terms of QRS morphology and location of early endocardial activation sites. This might be used to attain a better patient-specific knowledge of activation sequence from routine body-surface electrocardiograms.

Stability analysis of the POD reduced order method for solving the bidomain model in cardiac electrophysiology.

In this paper we show the numerical stability of the Proper Orthogonal Decomposition (POD) reduced order method used in cardiac electrophysiology applications. The difficulty of proving the stability comes from the fact that we are interested in the bidomain model, which is a system of degenerate parabolic equations coupled to a system of ODEs representing the cell membrane electrical activity. The proof of the stability of this method is based on a priori estimates controlling the gap between the reduced order solution and the Galerkin finite element one. We present some numerical simulations confirming the theoretical results. We also combine the POD method with a time splitting scheme allowing a faster solving of the bidomain problem and show numerical results. Finally, we conduct numerical simulation in 2D illustrating the stability of the POD method in its sensitivity to the ionic model parameters. We also perform 3D simulation using a massively parallel code. We show the computational gain using the POD reduced order model. We also show that this method has a better scalability than the full finite element method.

Effects of L-type calcium channel and human ether-a-go-go related gene blockers on the electrical activity of the human heart: a simulation study.

AIMS: Class III and IV drugs affect cardiac human ether-a-go-go related gene (IKr) and L-type calcium (ICaL) channels, resulting in complex alterations in repolarization with both anti- and pro-arrhythmic consequences. Interpretation of their effects on cellular and electrocardiogram (ECG)-based biomarkers for risk stratification is challenging. As pharmaceutical compounds often exhibit multiple ion channel effects, our goal is to investigate the simultaneous effect of ICaL and IKr block on human ventricular electrophysiology from ionic to ECG level. METHODS AND RESULTS: Simulations are conducted using a human body torso bidomain model, which includes realistic representation of human membrane kinetics, anatomy, and fibre orientation. A simple block pore model is incorporated to simulate drug-induced ICaL and IKr blocks, for drug dose = 0, IC50, 2× IC50, 10× IC50, and 30× IC50. Drug effects on human ventricular activity are quantified for different degrees and combinations of ICaL and IKr blocks from the ionic to the body surface ECG level. Electrocardiogram simulations show that ICaL block results in shortening of the QT interval, ST elevation, and reduced T-wave amplitude, caused by reduction in action potential duration and action potential amplitude during the plateau phase, and in repolarization times. In contrast, IKr block results in QT prolongation and reduced T-wave amplitude. When ICaL and IKr blocks are combined, the degree of ICaL block strongly determines QT interval whereas the effect of IKr block is more pronounced on the T-wave amplitude. CONCLUSION: Our simulation study provides new insights into the combined effect of ICaL and IKr blocks on human ventricular activity using a multiscale computational human torso model.

Chaste: an open source C++ library for computational physiology and biology.

Chaste - Cancer, Heart And Soft Tissue Environment - is an open source C++ library for the computational simulation of mathematical models developed for physiology and biology. Code development has been driven by two initial applications: cardiac electrophysiology and cancer development. A large number of cardiac electrophysiology studies have been enabled and performed, including high-performance computational investigations of defibrillation on realistic human cardiac geometries. New models for the initiation and growth of tumours have been developed. In particular, cell-based simulations have provided novel insight into the role of stem cells in the colorectal crypt. Chaste is constantly evolving and is now being applied to a far wider range of problems. The code provides modules for handling common scientific computing components, such as meshes and solvers for ordinary and partial differential equations (ODEs/PDEs). Re-use of these components avoids the need for researchers to 're-invent the wheel' with each new project, accelerating the rate of progress in new applications. Chaste is developed using industrially-derived techniques, in particular test-driven development, to ensure code quality, re-use and reliability. In this article we provide examples that illustrate the types of problems Chaste can be used to solve, which can be run on a desktop computer. We highlight some scientific studies that have used or are using Chaste, and the insights they have provided. The source code, both for specific releases and the development version, is available to download under an open source Berkeley Software Distribution (BSD) licence at http://www.cs.ox.ac.uk/chaste, together with details of a mailing list and links to documentation and tutorials.

Computational assessment of drug-induced effects on the electrocardiogram: from ion channel to body surface potentials.

BACKGROUND AND PURPOSE: Understanding drug effects on the heart is key to safety pharmacology assessment and anti-arrhythmic therapy development. Here our goal is to demonstrate the ability of computational models to simulate the effect of drug action on the electrical activity of the heart, at the level of the ion-channel, cell, heart and ECG body surface potential. EXPERIMENTAL APPROACH: We use the state-of-the-art mathematical models governing the electrical activity of the heart. A drug model is introduced using an ion channel conductance block for the hERG and fast sodium channels, depending on the IC(50) value and the drug dose. We simulate the ECG measurements at the body surface and compare biomarkers under different drug actions. KEY RESULTS: Introducing a 50% hERG-channel current block results in 8% prolongation of the APD(90) and 6% QT interval prolongation, hERG block does not affect the QRS interval. Introducing 50% fast sodium current block prolongs the QRS and the QT intervals by 12% and 5% respectively, and delays activation times, whereas APD(90) is not affected. CONCLUSIONS AND IMPLICATIONS: Both potassium and sodium blocks prolong the QT interval, but the underlying mechanism is different: for potassium it is due to APD prolongation; while for sodium it is due to a reduction of electrical wave velocity. This study shows the applicability of in silico models for the investigation of drug effects on the heart, from the ion channel to the ECG-based biomarkers.

A mathematical model of the Purkinje-muscle junctions.

This paper is devoted to the construction of a mathematical model of the His-Purkinje tree and the Purkinje-Muscle Junctions (PMJ). A simple numerical scheme is proposed in order to perform some simple numerical experiments.