Premature Ventricular Contraction Beats Research Articles

Design of a Low-area Hardware Architecture to Predict Early Signs of Sudden Cardiac Arrests

Sudden cardiac arrest (SCA) results in an unexpected and untimely death within minutes, and its early prediction can alert cardiac patients to a timely medical diagnosis. To detect early symptoms of an SCA, the detection and classification of ventricular tachycardias (VT) are of utmost importance. In this work, a low-area yet highly accurate hardware architecture for VT classification is proposed based on the detection of premature ventricular contraction (PVC) beats. After pre-processing of the ECG signals using a wavelet-based pre-processing unit, a characteristics-matching algorithm is used to detect the PVC beats, and a low-complexity adaptive decision-based logic classifier is used to classify them into four types of VTs, namely monomorphic, polymorphic, non-sustained VT (NSVT), and sustained VT (SVT). FPGA verification of the hardware architecture for the VT classifier using the Nexys 4 DDR Artix-7 board utilizes 10.4% of the total available resources and displays the type of VT and the number of PVCs detected to help in determining the severity of SCA and the need for medical attention. The ASIC implementation of the proposed PVC-based VT classification using the SCL 180 nm CMOS technology results in an area overhead of 0.02 mm2 and a power consumption of 3.47 μW for a high accuracy rate of 98.2%. When compared to the existing CA detection systems for wearable devices, the proposed one consumes the least area while achieving high detection rates.

Abstract 15352: Deep Learning Model-Enabled Electrocardiogram to Localize Premature Ventricular Contractions in Patients Referred for Catheter Ablation

Introduction: Accurate pre-procedure localization of the premature ventricular contractions (PVCs) is critical in preparing for the ablation. So far, no tools exist for artificial intelligence guided PVC-localization. Hypothesis: Trained deep learning model will be able to predict the site of origin (SOO) of PVCs using 12 lead electrocardiogram (ECG). Methods: We analyzed 707 ECG-ablation pairs from 2403 patients who had at least one ECG flagged for PVCs in the 6 months prior to their procedure. First, we used a cohort of 100,000 ECGs without PVCs and 10,000 ECGs with PVCs to train a convolutional neural network (CNN) to isolate the ECG segments with PVCs in them from a 12-lead ECG (Figure 1). We then trained two models, one utilizing a 10-second 12-lead ECG, and the other with just the isolated PVC beats as input. The models were trained using the Keras framework with TensorFlow (Google). Results: The mean age of the patients in the cohort was 61.2 ± 15.0 years. 66.3 percent of patients underwent left ventricle ablation, 54.6 percent had outflow ablation, and 55.3 percent had ablation of sites with leftward axis (septal location for midline and left ventricle PVCs). The model trained on 10-second ECGs yielded validation and test AUCs of 0.941 and 0.931 for outflow, 0.898 and 0.891 for ventricle, and 0.863 and 0.846 for leftward axis (Figure 1A). The model trained on isolated segments containing PVC beats yielded validation and test AUCs of 0.966 and 0.957 for outflow, 0.879 and 0.907 for ventricle, and 0.846 and 0.829 for SOO for leftward axis (Figure 1B). Conclusions: We have trained an accurate convolutional neural network that can predict SOO using the surface ECG. Further fine-tuning would be needed to refine the model and develop prediction of specific sites.

Electrocardiogram morphological arrhythmia classification using fuzzy entropy-based feature selection and optimal classifier

Electrocardiogram (ECG) signal analysis has become significant in recent years as cardiac arrhythmia shares a major portion of all mortality worldwide. To detect these arrhythmias, computer-assisted algorithms play a pivotal role as beat-by-beat monitoring of holter ECG signals is required. In this paper, a morphological arrhythmia classification algorithm has been proposed to classify seven different ECG beats, namely Normal Beat (N), Left Bundle Branch Block Beat (L), Right Bundle Branch Block Beat (R), Atrial Premature Contraction Beat (A), Premature Ventricular Contraction Beat (V), Fusion of Normal and Ventricle Beat (F) and Pace Beat (P). A novel feature set of 25 attributes has been extracted from each ECG beat and ranked using the Fuzzy Entropy-based feature selection (FEBFS) technique. In addition, two distinct classifiers, support vector machine with radial basis function as the kernel (SVM-RBF) and weighted K-nearest neighbor (WKNN), are used to categorize ECG beats, and their performances are also evaluated after adjusting vital parameters. The performance of classifiers is compared for four different ECG beat segmentation approaches and further analyzed using three similarity measurement techniques and two fuzzy entropy methods while feature selection. The classifier results are also cross-validated using a 10-fold cross-validation scheme, and the MIT-BIH Arrhythmia Database has been used to validate the proposed work. After selecting 21 highly ranked features, WKNN achieves the best results with the nearest neighbor value K = 3 and cityblock distance metrics, with Average Sensitivity (Sen) = 94.89%, Positive Predictivity (Ppre) = 97.13%, Specificity (Spe) = 99.72%, F1 Score = 95.95%, and Overall Accuracy (Acc) = 99.15%. The novelty of this work relies on formulating a unique feature set, including proposed symbolic features, followed by the FEBFS technique making this algorithm efficient and reliable for morphological arrhythmia classification. The above results demonstrate that the proposed algorithm performs better than many existing state-of-the-art works.

Design of hardware-efficient PVC recognition and classification system for early detection of sudden cardiac arrests

Sudden cardiac arrest (SCA) is a critical cardiovascular condition that needs greater emphasis on its early detection because it occurs unexpectedly and is fatal within minutes. Nowadays, wearable healthcare technologies are preferred for continuous monitoring of a wide range of cardiovascular diseases. Additionally, the recognition of cardiac arrhythmias such as premature ventricular contraction (PVC) and non-sustained ventricular tachycardia (NSVT) by wearable healthcare devices can help in the early detection of SCAs. The biggest challenge in this regard for compactly sized wearable healthcare devices is to design a hardware-efficient yet highly accurate PVC detection and classification system. In this work, a novel and hardware-efficient PVC recognition and classification system capable of categorizing PVCs into five types of ventricular arrhythmias, including NSVT, for the early detection of SCAs is proposed. For PVC detection, lifting-based discrete wavelet transform (LDWT) is used to perform pre-processing and extract morphological electrocardiogram (ECG) features and the Teager energy operator. When compared to the state-of-the-art ventricular arrhythmias and PVC detection systems, the proposed system performs PVC beat recognition and classification with an improved accuracy and sensitivity of 98.29% and 98.64%, respectively, while utilizing only 4.36% of the total resources when implemented on the Nexys 4 DDR FPGA board.

An intelligent diagnostic method of ECG signal based on Markov transition field and a ResNet

Background and ObjectiveHeart disease seriously threatens human life and health. It has the character of abruptness and is necessary to accurately monitor and intelligently diagnose electrocardiograph signals in real-time. As part of the automation of heart monitoring, the electrocardiogram (ECG) intelligent diagnosis method based on deep learning not only meets the needs of real-time and accurate but also can abandon relevant professional knowledge, which makes it possible to be promoted in the general population. MethodsThis paper presents an intelligent diagnosis method based on a ResNet. Firstly, ECG signals from MIT-BIH Database are converted into 2-dim matrices by Markov Transition Field. Secondly, the matrices are used as the input of a ResNet. Then, the ResNet is able to extract high abstract features of various diseases and realize intelligent identification of five heartbeat types, including Normal Beat, Left Bundle Branch Block Beat, Right Bundle Branch Block Beat, Premature Ventricular Contraction Beat, and Atrial Premature Contraction Beat. Eventually, the proposed model is used to identify Normal Beat and Atrial Fibrillation(AF) based on the PAF Prediction Challenge Database(the PAFPC Database) to verify its generalization ability. ResultsThe experiment result shows that the intelligent diagnosis method can reach a high F1-score of 97.7% and a high accuracy upon to 99.2% on MIT-BIH Database, which are higher than the models proposed by other researchers. Its mean sensitivity and mean specificity are 97.42% and 99.54%, respectively. Moreover, the accuracy of the generalization ability verification experiment is 94.57% on the PAFPC Database, which is also higher than the results of other studies. ConclusionThe research results show that the method proposed in this paper still achieves higher accuracy and higher F1-score than other methods without any data preprocessing. This method has better classification performance than traditional machine learning methods and other deep learning methods. That is, the method based on Markov Transition Field and a ResNet has good application prospects. At the same time, it has been verified that the model proposed in this paper also has excellent generalization ability.

Comparison of dipole-based and potential-based ECGI methods for premature ventricular contraction beat localization with clinical data.

Introduction: Localization of premature ventricular contraction (PVC) origin to guide the radiofrequency ablation (RFA) procedure is one of the prominent clinical goals of non-invasive electrocardiographic imaging. However, the results reported in the literature vary significantly depending on the source model and the level of complexity in the forward model. This study aims to compare the paced and spontaneous PVC localization performances of dipole-based and potential-based source models and corresponding inverse methods using the same clinical data and to evaluate the effects of torso inhomogeneities on these performances. Methods: The publicly available EP solution data from the EDGAR data repository (BSPs from a maximum of 240 electrodes) with known pacing locations and the Bratislava data (BSPs in 128 leads) with spontaneous PVCs from patients who underwent successful RFA procedures were used. Homogeneous and inhomogeneous torso models and corresponding forward problem solutions were used to relate sources on the closed epicardial and epicardial-endocardial surfaces. The localization error (LE) between the true and estimated pacing site/PVC origin was evaluated. Results: For paced data, the median LE values were 25.2 and 13.9mm for the dipole-based and potential-based models, respectively. These median LE values were higher for the spontaneous PVC data: 30.2-33.0mm for the dipole-based model and 28.9-39.2mm for the potential-based model. The assumption of inhomogeneities in the torso model did not change the dipole-based solutions much, but using an inhomogeneous model improved the potential-based solutions on the epicardial-endocardial ventricular surface. Conclusion: For the specific task of localization of pacing site/PVC origin, the dipole-based source model is more stable and robust than the potential-based source model. The torso inhomogeneities affect the performances of PVC origin localization in each source model differently. Hence, care must be taken in generating patient-specific geometric and forward models depending on the source model representation used in electrocardiographic imaging (ECGI).

A "two-step classification" machine learning method for non-invasive localization of premature ventricular contraction origins based on 12-lead ECG.

Premature ventricular contraction (PVC) is a type of cardiac arrhythmia that originates from ectopic pacemaker in the ventricles. The localization of the origin of PVC is essential for successful catheter ablation. However, most studies on non-invasive PVC localization focus on elaborate localization in specific regions of the ventricle. This study aims to propose a machine learning algorithm based on 12-lead electrocardiogram (ECG) data that can improve the accuracy of PVC localization in the whole ventricle. We collected 12-lead ECG data from 249 patients with spontaneous or pacing-induced PVCs. The ventricle was divided into 11 segments. In this paper, we propose a machine learning method consisting of two consecutive classification steps. In the first classification step, each PVC beat was labeled to one of the 11 ventricular segments using six features, including a newly proposed morphological feature called "Peak_index." Four machine learning methods were tested for comparative multi-classification performance and the best classifier result was kept to the next step. In the second classification step, a binary classifier was trained using a smaller combination of features to further differentiate segments that are easily confused. The Peak_index as a proposed new classification feature combined with other features is suitable for whole ventricle classification by machine learning methods. The test accuracy of the first classification reached 75.87%. It is shown that a second classification for confusable categories can improve the classification results. After the second classification, the test accuracy reached 76.84%, and when a sample classified into adjacent segments was considered correct, the test "rank accuracy" was improved to 93.49%. The binary classification corrected 10% of the confused samples. This paper proposes a "two-step classification" method to localize the origin of PVC beats into the 11 regions of the ventricle using non-invasive 12-lead ECG. It is expected to be a promising technique to be used in clinical settings to help guide ablation procedures.

An Automatic Premature Ventricular Contraction Recognition System Based on Imbalanced Dataset and Pre-Trained Residual Network Using Transfer Learning on ECG Signal

The development of automatic monitoring and diagnosis systems for cardiac patients over the internet has been facilitated by recent advancements in wearable sensor devices from electrocardiographs (ECGs), which need the use of patient-specific approaches. Premature ventricular contraction (PVC) is a common chronic cardiovascular disease that can cause conditions that are potentially fatal. Therefore, for the diagnosis of likely heart failure, precise PVC detection from ECGs is crucial. In the clinical settings, cardiologists typically employ long-term ECGs as a tool to identify PVCs, where a cardiologist must put in a lot of time and effort to appropriately assess the long-term ECGs which is time consuming and cumbersome. By addressing these issues, we have investigated a deep learning method with a pre-trained deep residual network, ResNet-18, to identify PVCs automatically using transfer learning mechanism. Herein, features are extracted by the inner layers of the network automatically compared to hand-crafted feature extraction methods. Transfer learning mechanism handles the difficulties of required large volume of training data for a deep model. The pre-trained model is evaluated on the Massachusetts Institute of Technology-Beth Israel Hospital (MIT-BIH) Arrhythmia and Institute of Cardiological Technics (INCART) datasets. First, we used the Pan-Tompkins algorithm to segment 44,103 normal and 6423 PVC beats, as well as 106,239 normal and 9987 PVC beats from the MIT-BIH Arrhythmia and IN-CART datasets, respectively. The pre-trained model employed the segmented beats as input after being converted into 2D (two-dimensional) images. The method is optimized with the using of weighted random samples, on-the-fly augmentation, Adam optimizer, and call back feature. The results from the proposed method demonstrate the satisfactory findings without the using of any complex pre-processing and feature extraction technique as well as design complexity of model. Using LOSOCV (leave one subject out cross-validation), the received accuracies on MIT-BIH and INCART are 99.93% and 99.77%, respectively, suppressing the state-of-the-art methods for PVC recognition on unseen data. This demonstrates the efficacy and generalizability of the proposed method on the imbalanced datasets. Due to the absence of device-specific (patient-specific) information at the evaluating stage on the target datasets in this study, the method might be used as a general approach to handle the situations in which ECG signals are obtained from different patients utilizing a variety of smart sensor devices.

A clue to the triples from an echo.

A clue to the triples from an echo.

Automated Detection of Abnormalities in ECG signals using Deep Neural Network

The electrocardiogram (ECG) is a diagnostic procedure that uses a skin electrode to record the heart's electrical activity. Heart diseases are the leading cause of mortality globally, and they have significant monetary cost. In the clinic, automated detection and classification technology of abnormalities in heart beats can assist physicians in making prompt and accurate medical diagnoses. This paper concentrates on two proposed models: Model A: 1-D 8-Layer CNN Model and Model B: 1- D CNN and the LSTM model to characterise ECG data into five categories: normal beat, right bundle branch block beat, left bundle branch block beat, premature ventricular contraction beat, and atrial premature beat. The data used to create and validate the models is obtained from the MIT-BIH Dataset, which comprises of 48 half-hour samples of two-lead continuous ECG signals obtained from 48 individuals. The dataset is divided into two files-.csv and.txt. For each sample, the.csv files include the readings collected from both leads. The experimental results show that Model A has an Accuracy = 99.68%, Precision = 99.23%, and a F1 score = 99.22%. And Model B has an Accuracy= 99.51%, Precision = 98.76%, and a F1 score = 98.76%. Our study aims to assist the medical sector by reducing the diagnoses time by automating the process of detecting arrythmias.

Flexible ECG signal modeling and compression using alpha stable functions

In this paper, we propose a flexible modeling technique for the ECG signals. Modeling is achieved by considering the weighted summation of elementary functions representing the waveforms that describe each component of the cardiac cycle. We thus evaluate the benefit brought by α-stable functions with respect to Gaussian functions in terms of modeling precision. Seven records from the MIT-BIH arrhythmia database have been used to assess the performances of the proposed modeling method, including Normal beat, Premature Ventricular Contraction beat, Right Bundle Block Branch beat and Paced beat. When applied on the chosen records, it turns out that α-stable modeling always outperforms Gaussian modeling. Since each waveform is related to a particular physiological event in the ECG cardiac cycle, we also exploit flexibility of choosing α-stable modeling instead of Gaussian one for some event of medical interest in order to solve compression purpose efficiency-quality compromise. The comparison of the α-stable model applied in compression with other techniques proves the efficiency of the proposed method, mainly in term of quality score.

Premature Ventricular Contractions Are Presaged by a Mechanically Abnormal Sinus Beat.

Frequent premature ventricular contractions (PVCs) can lead to cardiomyopathy; it is unclear if there are abnormal myocardial mechanics operative in the PVC and non-PVC beats. The aim of this study was to investigate regional and global myocardial mechanics, including dyssynchrony, in patients with frequent PVCs. Fifty-six consecutive patients referred for PVC ablation were prospectively studied. During sinus rhythm (SR) and PVC beats, left ventricular (LV) global longitudinal strain (GLS), LV dyssynchrony (measured as the SD of time to peak GLS), and dyssynergy (measured as maximum regional strain minus minimum regional strain at aortic valve closure) were quantified using 2-dimensional strain echocardiography. GLS, dyssynchrony, and dyssynergy were compared in remote SR, pre-PVC SR, PVC, and post-PVC SR beats. In SR beats remote from the PVC, GLS was-17.3% ± 4%, dyssynchrony was 49 ± 14ms, and dyssynergy was22% ± 9%. Myocardial mechanics were significantly abnormal during PVCs compared with remote SR beats (GLS-7.7% ± 3% [P< 0.001], dyssynchrony 115 ± 37 milliseconds [P< 0.001], and dyssynergy 26% ± 10% [P<0.001]). There were significant mechanical abnormalities in the SR beat preceding the PVC, which demonstrated significantly lower LV strain (pre-PVC SR,-13% ± 4%; P< 0.001) and more dyssynchrony (pre-PVC SR, 63 ± 19 milliseconds; P< 0.001) compared with remote SR beats. Dyssynergy was significantly higher for pre-PVC SR and PVC beats compared with remote SR (pre-PVC SR, 25% ± 8% [P< 0.001]; PVC, 26% ± 10% [P< 0.001]). In patients with frequent PVCs, the SR beat preceding the PVC demonstrates significant mechanical abnormalities. This finding suggests that perturbations in cellular physiological processes such as excitation-contraction coupling may underlie the generation of frequent PVCs.

Arrhythmia Assessment With 30-day MCT Patch Challenges Traditional Holter Monitoring

<h3>Introduction</h3> : Delayed diagnosis of episodic and asymptomatic arrhythmias can lead to adverse outcomes, especially for patients in the outpatient setting. Ambulatory Holter monitoring has been a cornerstone for diagnosing arrhythmias. While a traditional Holter records ECG signal for 24-48 hrs, the mobile cardiac telemetry (MCT) patch allows real-time ECG monitoring for up to 30 days. An MCT patch provides the ability to analyze long-term ECG data in real-time for timely and improved diagnosis. <h3>Objective</h3> : This study aims to identify if cardiac monitoring with a 30-day MCT patch provides better patient outcome than with the 24-hr Holter. To do so, we compared the arrhythmia and ectopy diagnostic yield of both modalities. <h3>Methods</h3> : A retrospective study was designed to analyze data recorded from 14,514 patients (male-43%, female-57%, median age-64 years) who were prescribed to be monitored with Zywie ECG monitors (Zywie, Inc., Georgia). The mean ECG analyzable time from 12,974 patients who used Holter and 1,540 patients who used MCT patch was 18.60±1.07 hrs and 28.2±2.4 days, respectively. We evaluated data to find the incidences of ectopic beats, including premature atrial contraction (PAC) and premature ventricular contraction (PVC), and any of 5 arrhythmias, including, sinus bradycardia (SB), sinus tachycardia (ST), atrial fibrillation (AF), pause, and ventricular tachycardia (VT). <h3>Results</h3> : Fig.1 suggests that the mean arrhythmia diagnostic yield of 38.1% in 30-day MCT studies is significantly different from 29.1% in 24-hr Holter studies (<i>p< 0.001</i>). We also analyzed the dataset to identify patients in which at least one arrhythmia was identified excluding SB and ST events. We identified 25% critical arrhythmias (excluding SB and ST) in patients with MCT patches in contrast to 9.6% in Holter. Fig. 2 shows that PVC beats were identified in 85.9% and 65.7%, whereas PAC beats in 92.3% and 78.2% of MCT and Holter studies, respectively. <h3>Conclusion</h3> : Results demonstrate that 30-day monitoring with MCT patch provides a better diagnostic yield than 24-hr Holter monitoring. Timely diagnosis with MCT patch can aid in early therapeutic intervention resulting in better patient care.

A comparative study and analysis of LSTM deep neural networks for heartbeats classification

Heart diseases and their diagnosis has become a predominant topic in Healthcare systems as the heart is one of the pivotal parts of the human body. Electrocardiogram (ECG) signal-based diagnosis and classification have been experimented with various computational techniques which have demonstrated early detection and treatment of heart disease. Deep learning (DL) is the current interest of different Healthcare applications that includes the heartbeat classification based on ECG signals. There are various studies conducted with different DL models, such as Convolutional Neural Network (CNN), Recurrent Neural Network (RNN) and Gated Recurrent Unit (GRU) for the heartbeat classification using MIT-BIH arrhythmia dataset. This paper aims to provide a comprehensive analysis of Long-Short Term Memory (LSTM) based DL models with multiple performance metrics on the MIT-BIH arrhythmia dataset for the heartbeat classification. The different variants of the LSTM DL model are proposed for the purpose of the classification. Among the variants, the bi-directional LSTM DL model shows high accuracy in the classification of Normal beats (97%), Premature ventricular contractions (PVC) beats (98%), Atrial Premature Complex (APC) beats (98%), and Paced Beats (PB) beats (99%). The comparative analysis of the bi-directional LSTM DL model with the existing works shows 95% sensitivity and 98% specificity in the classification of heartbeats. The results evidently show that the LSTM DL models are appropriate for the classification of heartbeats.

Machine learning-data mining integrated approach for premature ventricular contraction prediction

Cardiac arrhythmias impose a significant burden on the healthcare environment due to the increasing ratio of mortality worldwide. Arrhythmia and abnormal ECG heartbeat are the possible symptoms of severe heart diseases that can lead to death. Premature ventricular contraction (PVC) is a common form of cardiac arrhythmia which begins from the lower chamber of the heart, and frequent occurrence of PVC beat might lead to mortality. ECG signals are the noninvasive and primary tool used to identify the actual life threat related to the heart. Nowadays, in society, the computer-assisted technique reduces doctors' burden to evaluate heart disease and heart arrhythmia automatically. Regardless of well-equipped and well-developed health facilities that are available for monitoring the cardiac condition, the success stories are yet unsatisfactorily due to the complexity of the cardiac disorder. The most challenging part in ECG signal analysis is to extract the accurate features relevant to the arrhythmia for classification due to the inter-patient variation. There are many morphological changes present in the ECG signals. Hence, there is a gap in the usage of appropriate methods for the extraction of features and classification models, which reduce the biased diagnosis of PVC arrhythmia. To predict PVC arrhythmia accurately is a quite challenging task owing to (a) QRS negative (b) long compensatory pause (c) p-wave (d) biased diagnosis of PVC detection due to the small feature set. This study presents a new approach for PVC prediction using derived predictor variables from the electrocardiograph (ECG-MLII) signals: R–R wave interval, previous R–R wave interval, QRS duration, and verification of P-wave whether it is present or absent using threshold technique. We propose the machine learning-data mining MACDM integrated approach using five different models of multiple logistic regression and four classifiers, namely, Random Forest (RF), K-Nearest Neighbor (KNN), Support vector machine (SVM), and Naive Bayes (NB). The experiment was conducted on the public benchmark MIT-BIH-AR to evaluate the performance of our proposed MACDM technique. The multiple logistic regression models constructed as a function of all independent variables achieved an accuracy of 99.96%, sensitivity 98.9%, specificity 99.20%, PPV 99.25%, and Youden's index parameter 98.24%. Thus, it is proved that this computer-aided method helps our medical practitioners improve the efficiency of their services.

PVC arrhythmia classification based on fractional order system modeling.

It is well known that many physiological phenomena are modeled accurately and effectively using fractional operators and systems. This type of modeling is due mainly to the dynamical link between fractional-order systems and the fractal structures of the physiological systems. The automatic characterization of the premature ventricular contraction (PVC) is very important for early diagnosis of patients with different life-threatening cardiac diseases. In this paper, a classification scheme of normal and PVC beats of the electrocardiogram (ECG) signal is proposed. The clustering features used for normal and PVC beats discrimination are the parameters of the commensurate order linear fractional model of the frequency content of the QRS complex of the ECG signal. A series of tests and comparisons have been performed to evaluate and validate the efficiency of the proposed PVC classification algorithm using the MIT-BIH arrhythmia database. The proposed PVC classification method has achieved an overall accuracy of 94.745%, a specificity of 95.178% and a sensitivity of 90.021% using all the 48 records of the database.

Robust deep learning pipeline for PVC beats localization.

BACKGROUND: Premature ventricular contraction (PVC) is among the most frequently occurring types of arrhythmias. Existing approaches for automated PVC identification suffer from a range of disadvantages related to hand-crafted features and benchmarking on datasets with a tiny sample of PVC beats.OBJECTIVE: The main objective is to address the drawbacks described above in the proposed framework, which takes a raw ECG signal as an input and localizes R peaks of the PVC beats.METHODS: Our method consists of two neural networks. First, an encoder-decoder architecture trained on PVC-rich dataset localizes the R peak of both Normal and anomalous heartbeats. Provided R peaks positions, our CardioIncNet model does the delineation of healthy versus PVC beats.RESULTS: We have performed an extensive evaluation of our pipeline with both single- and cross-dataset paradigms on three public datasets. Our approach results in over 0.99 and 0.979 F1-measure on both single- and cross-dataset paradigms for R peaks localization task and above 0.96 and 0.85 F1 score for the PVC beats classification task.CONCLUSIONS: We have shown a method that provides robust performance beyond the beats of Normal nature and clearly outperforms classical algorithms both in the case of a single and cross-dataset evaluation. We provide a Github repository for the reproduction of the results.

Premature Ventricular Contractions and Ultra-High-Definition Mapping. Contribution and Limits.

The utility of ultra-high definition mapping (UHDM) for ablation of premature ventricular contractions (PVC) remains undetermined. The aim of this study was to investigate UHDM for PVC ablation, and additionally to compare to conventional technique. Twenty patients investigated using UHDM were prospectively included and analyzed. Electrophysiological caracteristics and results were compared to 40 patients ablated using fluoroscopy only. 2541±2033 EGMs and 331±240 PVC beats were recorded for each patient. Surfaces of isochronal activations were 2.3±1.7 and 6.9±6.1 cm2 (first 10 and 20 ms). Local scar was present in 40% and local block in 65%. Areas of pace-mapping > 95, 90 and 85% concordance were 1.5±3.4, 2.1±3.9 and 3.3±5 cm2. Mean distance between the ablation site and the site of best pace-mapping or of earliest activation was 8±8 mm and 5±7 mm. Pre-potential was noted in 17% vs 26% controls (ns). QS pattern was present in 83% vs 83% controls (ns), and earliest activation was - 31±50 vs - 25±14 ms in controls (ns). Procedure (100±36 vs 190±51 min, p< 0.0001) and fluoroscopy duration (15±9 vs 24±9 min, p=0.005) were shorter in controls. Acute success was achieved in 65% patients with UHDM and in 72% controls (p=ns) with lower residual PVC burden in the control group. Over a follow-up of 19±12 months, long-term success was similar between groups (65 vs 68%). UHDM may reveal poorly recognized activation features and PVC mechanism. In this series, conventional mapping was quicker and did clinically as well as UHDM.

Recognition of Premature Ventricular Contraction Beat from 12Lead ECG Based on A Novel Detection Function of QRS Onset.

Premature ventricular contraction (PVC) is associated to the risk of ventricular dysfunction and cardiovascular events. Its diagnosis depends on a long-time monitoring, and computational tools for PVC recognition can provide significant assistance to specialists. For this purpose, we present an automatic algorithm for the recognition PVC beat based on long-term 12-lead ECG.A total of 249 patients with PVC were included in this study. Initially, a novel QRS onset detection function was used to automatically extract QRS complexes from massive original ECG data. Then, non-personalized but shared QRS-width features of 12-lead QRS complexes were extracted and fed to a binary classifier based on SVM. In order to verify the model, 17, 512 normal beats and 17, 690 PVC beats extracted from 35 patients were used for training, and another 215 normal beats and 291 PVC beats selected randomly from the remaining 214 patients were used for testing.As a result, the achieved accuracy, sensitivity, specificity in training data and testing data are 98.9%, 98.3%, 99.5% and 97.2%, 97.7%, 96.7%, respectively. The high accuracy of PVC recognition makes it promising to be an efficient technique being used in clinical settings to automatically analyze huge ECG data so as to replace the tedious manual interpretation.

Classification of Arrhythmia by Using Deep Learning with 2-D ECG Spectral Image Representation

The electrocardiogram (ECG) is one of the most extensively employed signals used in the diagnosis and prediction of cardiovascular diseases (CVDs). The ECG signals can capture the heart’s rhythmic irregularities, commonly known as arrhythmias. A careful study of ECG signals is crucial for precise diagnoses of patients’ acute and chronic heart conditions. In this study, we propose a two-dimensional (2-D) convolutional neural network (CNN) model for the classification of ECG signals into eight classes; namely, normal beat, premature ventricular contraction beat, paced beat, right bundle branch block beat, left bundle branch block beat, atrial premature contraction beat, ventricular flutter wave beat, and ventricular escape beat. The one-dimensional ECG time series signals are transformed into 2-D spectrograms through short-time Fourier transform. The 2-D CNN model consisting of four convolutional layers and four pooling layers is designed for extracting robust features from the input spectrograms. Our proposed methodology is evaluated on a publicly available MIT-BIH arrhythmia dataset. We achieved a state-of-the-art average classification accuracy of 99.11%, which is better than those of recently reported results in classifying similar types of arrhythmias. The performance is significant in other indices as well, including sensitivity and specificity, which indicates the success of the proposed method.