Search

Early defibrillation by an automated external defibrillator (AED) is key for the survival of out-of-hospital cardiac arrest (OHCA) patients. ECG feature extraction and machine learning have been successfully used to detect ventricular fibrillation (VF) in AED shock decision algorithms. Recently, deep learning architectures based on 1D Convolutional Neural Networks (CNN) have been proposed for this task. This study introduces a deep learning architecture based on 1D-CNN layers and a Long Short-Term Memory (LSTM) network for the detection of VF. Two datasets were used, one from public repositories of Holter recordings captured at the onset of the arrhythmia, and a second from OHCA patients obtained minutes after the onset of the arrest. Data was partitioned patient-wise into training (80%) to design the classifiers, and test (20%) to report the results. The proposed architecture was compared to 1D-CNN only deep learners, and to a classical approach based on VF-detection features and a support vector machine (SVM) classifier. The algorithms were evaluated in terms of balanced accuracy (BAC), the unweighted mean of the sensitivity (Se) and specificity (Sp). The BAC, Se, and Sp of the architecture for 4-s ECG segments was 99.3%, 99.7%, and 98.9% for the public data, and 98.0%, 99.2%, and 96.7% for OHCA data. The proposed architecture outperformed all other classifiers by at least 0.3-points in BAC in the public data, and by 2.2-points in the OHCA data. The architecture met the 95% Sp and 90% Se requirements of the American Heart Association in both datasets for segment lengths as short as 3-s. This is, to the best of our knowledge, the most accurate VF detection algorithm to date, especially on OHCA data, and it would enable an accurate shock no shock diagnosis in a very short time., Competing Interests: The funders, Tecnalia Research and Innovation and BBVA, provided support in the form of salaries for authors AP, AA, FAA, CF, EG. This does not alter our adherence to PLOS ONE policies on sharing data and materials.

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

Early recognition of ventricular fibrillation (VF) and electrical therapy are key for the survival of out-of-hospital cardiac arrest (OHCA) patients treated with automated external defibrillators (AED). AED algorithms for VF-detection are customarily assessed using Holter recordings from public electrocardiogram (ECG) databases, which may be different from the ECG seen during OHCA events. This study evaluates VF-detection using data from both OHCA patients and public Holter recordings. ECG-segments of 4-s and 8-s duration were analyzed. For each segment 30 features were computed and fed to state of the art machine learning (ML) algorithms. ML-algorithms with built-in feature selection capabilities were used to determine the optimal feature subsets for both databases. Patient-wise bootstrap techniques were used to evaluate algorithm performance in terms of sensitivity (Se), specificity (Sp) and balanced error rate (BER). Performance was significantly better for public data with a mean Se of 96.6%, Sp of 98.8% and BER 2.2% compared to a mean Se of 94.7%, Sp of 96.5% and BER 4.4% for OHCA data. OHCA data required two times more features than the data from public databases for an accurate detection (6 vs 3). No significant differences in performance were found for different segment lengths, the BER differences were below 0.5-points in all cases. Our results show that VF-detection is more challenging for OHCA data than for data from public databases, and that accurate VF-detection is possible with segments as short as 4-s.

Background: Fast and accurate quality estimation of the electrocardiogram (ECG) signal is a relevant research topic that has attracted considerable interest in the scientific community, particularly due to its impact on tele-medicine monitoring systems, where the ECG is collected by untrained technicians. In recent years, a number of studies have addressed this topic, showing poor performance in discriminating between clinically acceptable and unacceptable ECG records., Methods: This paper presents a novel, simple and accurate algorithm to estimate the quality of the 12-lead ECG by exploiting the structure of the cross-covariance matrix among different leads. Ideally, ECG signals from different leads should be highly correlated since they capture the same electrical activation process of the heart. However, in the presence of noise or artifacts the covariance among these signals will be affected. Eigenvalues of the ECG signals covariance matrix are fed into three different supervised binary classifiers., Results and Conclusion: The performance of these classifiers were evaluated using PhysioNet/CinC Challenge 2011 data. Our best quality classifier achieved an accuracy of 0.898 in the test set, while having a complexity well below the results of contestants who participated in the Challenge, thus making it suitable for implementation in current cellular devices.

Early detection of ventricular fibrillation (VF) and rapid ventricular tachycardia (VT) is crucial for the success of the defibrillation therapy. A wide variety of detection algorithms have been proposed based on temporal, spectral, or complexity parameters extracted from the ECG. However, these algorithms are mostly constructed by considering each parameter individually. In this study, we present a novel life-threatening arrhythmias detection algorithm that combines a number of previously proposed ECG parameters by using support vector machines classifiers. A total of 13 parameters were computed accounting for temporal (morphological), spectral, and complexity features of the ECG signal. A filter-type feature selection (FS) procedure was proposed to analyze the relevance of the computed parameters and how they affect the detection performance. The proposed methodology was evaluated in two different binary detection scenarios: shockable (FV plus VT) versus nonshockable arrhythmias, and VF versus nonVF rhythms, using the information contained in the medical imaging technology database, the Creighton University ventricular tachycardia database, and the ventricular arrhythmia database. sensitivity (SE) and specificity (SP) analysis on the out of sample test data showed values of SE=95%, SP=99%, and SE=92% , SP=97% in the case of shockable and VF scenarios, respectively. Our algorithm was benchmarked against individual detection schemes, significantly improving their performance. Our results demonstrate that the combination of ECG parameters using statistical learning algorithms improves the efficiency for the detection of life-threatening arrhythmias.

Background: The location of the myocardial infarction (MI) might modify the spectral characteristics of ventricular fibrillation (VF) in humans., Objective: To evaluate the effect of the location of the infarcted area on the spectral parameters of VF., Methods: Patients with chronic MI (29 anterior, 32 inferior) and induced VF during cardioverter defibrillator implant were retrospectively studied. Dominant frequency (f(d)), organization index (OI), and power of the harmonic peaks were calculated in the device-stored electrograms (EGM) during sinus rhythm (SR) and VF., Results: The f(d) of the VF was not affected by the left ventricular ejection fraction (LVEF) or the MI location (anterior: 4.54 +/- 0.74 Hz, inferior: 4.77 +/- 0.48 Hz, n.s.). The OI was also similar in both groups. However, in patients with inferior MIs, normalized peak power at f(d) was higher (118.3 +/- 18.5 vs 100.6 +/- 28.2, P < 0.01) and the normalized peak power of the harmonics was lower than in the anterior MI group. The analysis of EGM during SR showed similar results. The size of the necrotic area and its distance to the recording electrode might partially explain these results., Conclusion: In our series, the spectral characteristics of the EGMs during VF showed significant differences depending on the MI localization. A higher fraction of energy (in the low-frequency region) was seen in inferior MIs, whereas the peak power at the harmonics increased in anterior MIs. A similar effect was seen during SR and VF, suggesting that it is caused by local electrophysiology abnormalities induced by the MI rather than by different intrinsic characteristics of the VF.

The long QT syndrome type-3 (LQT3) is an inherited cardiac disorder caused by mutations in the sodium channel gene SCN5A. LQT3 has been associated with ventricular arrhythmias and sudden cardiac death, specially at low heart rates. Based on computer simulations and experimental investigations, analysis of the morphology of the Action Potential (AP) has shown that it undergoes early afterdepolarizations (EADs) and spontaneous discharges, which are thought to be the trigger for reentry like-activity. However, dynamic characteristics of cardiac tissue are also important factors of arrhythmia mechanisms. In this work, we propose a dynamical analysis of the LQT3 at cellular level. We use a detailed Markovian model of the DeltaKPQ mutation, which is associated with LQT3, and we study beat-to-beat AP Duration (APD) variations by using a long-term stimulation protocol. Compared to wild-type (WT) cells, DeltaKPQ mutant cells are found to develop APD alternans over a narrow range of stimulation frequencies. Moreover, the interval of frequency dependence of APD alternans is related to the degree of severity of the EADs present in the AP. In conclusion, dynamical analysis of paced cells is a useful approach to understand the mechanisms of rate dependent arrhythmias.

The present paper describes a study where effects of anterior myocardium on body surface potentials were investigated. The study combines numerical lead field analysis combined with cardiac automata model. Electric fields are calculated with finite difference method in a 3-D model of male thorax. The cardiac activation applied in the study is an ectopic beat originating in the apex. The correlations and mean differences between signal generated by anterior segments of left ventricle and signal generated by both ventricles were analysed for 117 leads. The results show that there are leads which have high correlation (>0.9) with low the relative mean difference (<0.2) between the signals generated by anterior segments and signals generated by whole ventricles. These electrode locations would be optimal to monitor the activation of anterior segments when ectopic beats originate in apex.

The analysis of intracardiac Electrograms (EGM) recorded by transvenous lead systems in Implantable Cardioverter Defibrillators (ICD) often entails assumptions on the scope of the lead system. Based on bioelectric signal modeling and on numerical analysis, we studied quantitatively the scope of unipolar and bipolar lead configurations in ICD. We defined the scope in terms of the Mean Square Difference (MSD) between EGM generated by the whole myocardium, and EGM generated by different families of regions within the myocardium. For unipolar and bipolar lead systems, simulations showed that the smallest myocardial region involving a given value of MSD is characterized by the highest measurement sensitivity. Furthermore, the scope in the ventricles was found to be an order of magnitude smaller for bipolar leads than for unipolar leads. Bioelectric signal modeling combined with numerical analysis constitutes a powerful method to study quantitatively the scope of transvenous lead systems.

Introduction and Objectives: Mathematical models of cardiac electrical activity may help to elucidate the electrophysiological mechanisms involved in the genesis of arrhythmias. The most realistic simulations are based on reaction-diffusion models and involve a considerable computational burden. The aim of this study was to develop a computer model of cardiac electrical activity able to simulate complex electrophysiological phenomena but free of the large computational demands required by other commonly used models., Material and Method: A cellular automata system was used to model the cardiac tissue. Each individual unit had several discrete states that changed according to simple rules as a function of the previous state and the state of the neighboring cells. Activation was considered as a probabilistic process and was adjusted using restitution curves. In contrast, repolarization was modeled as a deterministic phenomenon. Cell currents in the model were calculated with a prototypical action potential that allowed virtual monopolar and bipolar electrograms to be simulated at any point in space., Results: Reproducible flat activation fronts, propagation from a focal stimulus, and reentry processes that were stable and unstable in two dimensions (with their corresponding electrograms) were obtained. The model was particularly suitable for the simulation of the effects observed in curvilinear activation fronts. Fibrillatory conduction and stable rotors in two- and three-dimensional substrates were also obtained., Conclusions: The probabilistic cellular automata model was simple to implement and was not associated with a high computational burden. It provided a realistic simulation of complex phenomena of interest in electrophysiology.