Improvements in atrial fibrillation detection for real-time monitoring

Bed sensor ballistocardiogram for non-invasive detection of atrial fibrillation: a comprehensive clinical study

Objective.Atrial fibrillation (AFib) is a common cardiac arrhythmia associated with high morbidity and mortality, making early detection and continuous monitoring essential to prevent complications like stroke. This study explores the potential of using a ballistocardiogram (BCG) based bed sensor for the detection of AFib.Approach.We conducted a comprehensive clinical study with night hospital recordings from 116 patients, divided into 72 training and 44 test subjects. The study employs established methods such as autocorrelation to identify AFib from BCG signals. Spot and continuous Holter ECG were used as reference methods for AFib detection against which BCG rhythm classifications were compared.Results.Our findings demonstrate the potential of BCG-based AFib detection, achieving 94% accuracy on the training set using a rule-based method. Furthermore, the machine learning model trained with the training set achieved an AUROC score of 97% on the test set.Significance.This innovative approach shows promise for accurate, non-invasive, and continuous monitoring of AFib, contributing to improved patient care and outcomes, particularly in the context of home-based or hospital settings.

Research on atrial fibrillation diagnosis in electrocardiograms based on CLA-AF model

Aims

The electrocardiogram (ECG) is the primary method for diagnosing atrial fibrillation (AF), but interpreting ECGs can be time-consuming and labour-intensive, which deserves more exploration.

Methods and results

We collected ECG data from 6590 patients as YY2023, classified as Normal, AF, and Other. Convolutional Neural Network (CNN), bidirectional Long Short-Term Memory (BiLSTM), and Attention construct the AF recognition model CNN BiLSTM Attention-Atrial Fibrillation (CLA-AF). The generalization ability of the model is validated on public datasets CPSC2018, PhysioNet2017, and PTB-XL, and we explored the performance of oversampling, resampling, and hybrid datasets. Finally, additional PhysioNet2021 was added to validate the robustness and applicability in different clinical settings. We employed the SHapley Additive exPlanations (SHAP) method to interpret the model's predictions. The F1-score, Precision, and area under the ROC curve (AUC) of the CLA-AF model on YY2023 are 0.956, 0.970, and 1.00, respectively. Similarly, the AUC on CPSC2018, PhysioNet2017, and PTB-XL reached above 0.95, demonstrating its strong generalization ability. After oversampling PhysioNet2017, F1-score and Recall improved by 0.156 and 0.260. Generalization ability varied with sampling frequency. The model trained from the hybrid dataset has the most robust generalization ability, achieving an AUC of 0.96 or more. The AUC of PhysioNet2021 is 1.00, which proves the applicability of CLA-AF. The SHAP values visualization results demonstrate that the model's interpretation of AF aligns with the diagnostic criteria of AF.

Conclusion

The CLA-AF model demonstrates a high accuracy in recognizing AF from ECG, exhibiting remarkable applicability and robustness in diverse clinical settings.

A deep learning modular ECG approach for cardiologist assisted adjudication of atrial fibrillation and atrial flutter episodes

Background

Detection of atrial tachyarrhythmias (ATA) on long-term electrocardiogram (ECG) recordings is a prerequisite to reduce ATA-related adverse events. However, the burden of editing massive ECG data is not sustainable. Deep learning (DL) algorithms provide improved performances on resting ECG databases. However, results on long-term Holter recordings are scarce.

Objective

We aimed to build and evaluate a DL modular software using ECG features well known to cardiologists with a user interface that allows cardiologists to adjudicate the results and drive a second DL analysis.

Methods

Using a large (n = 187 recordings, 249,419 one-minute samples), beat-to-beat annotated, two-lead Holter database, we built a DL algorithm with a modular structure mimicking expert physician ECG interpretation to classify atrial rhythms. The DL network includes 3 modules (cardiac rhythm regularity, electrical atrial waveform, and raw voltage by time data) followed by a decision network and a long-term weighting factor. The algorithm was validated on an external database.

Results

F1 scores of our classifier were 99% for ATA detection, 95% for atrial fibrillation, and 90% for atrial flutter. Using the external Massachusetts Institute of Technology database, the classifier obtains an F1-score of 97% for the normal sinus rhythm class and 96% for the ATA class. Residual errors could be corrected by manual deactivation of 1 module in 7 of 15 of the recordings, with an accuracy < 90%.

Conclusion

A DL modular software using ECG features well known to cardiologists provided an excellent overall performance. Clinically significant residual errors were most often related to the classification of the atrial arrhythmia type (fibrillation vs flutter). The modular structure of the algorithm helped to edit and correct the artificial intelligence-based first-pass analysis and will provide a basis for explainability.

A Narrowband IoT Personal Sensor for Long-Term Heart Rate Monitoring and Atrial Fibrillation Detection

Long-term patient monitoring is required for detection of episodes of atrial fibrillation, one of the most widespread cardiac pathologies. Today, the most used non-invasive technique is Holter electrocardiographic (ECG) monitoring, which can often prove ineffective because of the short duration of recordings (e.g., one day). Other techniques such as photo-plethysmography are adopted by smartwatches for much longer duration monitoring, but this has the disadvantage of offering only intermittent measurements. This study proposes an Internet of Things (IoT) sensor that can provide a very long period of continuous monitoring. The sensor consists of an ECG-integrated Analog Front End (MAX30003), a microcontroller (STM32F401RE), and an IoT narrowband module (STEVAL-STMODLTE). The instantaneous heart rate is extracted from the ECG recording in real time. At intervals of two minutes, the sequence of inter-beat intervals is transmitted to an IoT cloud platform (ThingSpeak). Settled atrial fibrillation event recognition software runs on the cloud and generates alerts when it recognizes such arrhythmia. Performances of the proposed sensor were evaluated by generating analog ECG signals from a public dataset of ECG signals with atrial fibrillation episodes, the MIT-BIH Atrial Fibrillation Database, each recording lasting approximately 10 h. Software implementing the Lorentz algorithm, one of the best detectors of atrial fibrillation, was implemented on the cloud platform. The accuracy, sensitivity, and specificity in recognizing atrial fibrillation episodes of the proposed system was calculated by comparison with a cardiologist's reference data. Across all patients, the proposed method achieved an accuracy of 0.88, a sensitivity 0.71, and a specificity 0.99. The results obtained suggest that the developed system can continuously record and transmit heart rhythms effectively and efficiently and, in addition, offers considerable performance in recognizing atrial fibrillation episodes in real time.

A Study of R-R Interval Transition Matrix Features for Machine Learning Algorithms in AFib Detection

Atrial Fibrillation (AFib) is a heart condition that occurs when electrophysiological malformations within heart tissues cause the atria to lose coordination with the ventricles, resulting in "irregularly irregular" heartbeats. Because symptoms are subtle and unpredictable, AFib diagnosis is often difficult or delayed. One possible solution is to build a system which predicts AFib based on the variability of R-R intervals (the distances between two R-peaks). This research aims to incorporate the transition matrix as a novel measure of R-R variability, while combining three segmentation schemes and two feature importance measures to systematically analyze the significance of individual features. The MIT-BIH dataset was first divided into three segmentation schemes, consisting of 5-s, 10-s, and 25-s subsets. In total, 21 various features, including the transition matrix features, were extracted from these subsets and used for the training of 11 machine learning classifiers. Next, permutation importance and tree-based feature importance calculations determined the most predictive features for each model. In summary, with Leave-One-Person-Out Cross Validation, classifiers under the 25-s segmentation scheme produced the best accuracies; specifically, Gradient Boosting (96.08%), Light Gradient Boosting (96.11%), and Extreme Gradient Boosting (96.30%). Among eleven classifiers, the three gradient boosting models and Random Forest exhibited the highest overall performance across all segmentation schemes. Moreover, the permutation and tree-based importance results demonstrated that the transition matrix features were most significant with longer subset lengths.

12-lead ECG signal processing and atrial fibrillation prediction in clinical practice

BACKGROUND

Because clinically used 12-lead electrocardiography (ECG) devices have high falsepositive errors in automatic interpretations of atrial fibrillation (AF), they require substantial improvements before use.

OBJECTIVE

A clinical 12-lead ECG pre-processing method with a parallel convolutional neural network (CNN) model for 12-lead ECG automatic AF recognition is introduced.

METHODS

Raw AF diagnosis data from a 12-lead ECG device were collected and analyzed by two cardiologists to differentiate between true- and false-positives. Using a stationary wavelet transform (SWT) and independent component analysis (ICA) noise reduction was conducted and baseline wandering was corrected for the raw signals. AF patterns were learned and predicted using a parallel CNN deep learning (DL) model. (1) The proposed method alleviates the decreased ECG QRS amplitude enhances the signal-to-noise ratio and clearly shows atrial and ventricular activities. (2) After training, the CNNbased AF detector significantly reduced false-positive errors. The precision of AF diagnosis increased from 77.3% to 94.0 ± 1.5% as compared to ECG device interpretation. For AF screening, the model showed an average sensitivity of 96.8 ± 2.2%, specificity of 79.0 ± 5.8%, precision of 94.0 ± 1.5%, F1-measure of 95.2 ± 1.0%, and overall accuracy of 92.7 ± 1.5%.

CONCLUSIONS

The method can bridge the gap between the research and clinical practice The ECG signal pre-processing and DL-based AF interpretation can be rapidly implemented clinically.

A comparative study on neural networks for paroxysmal atrial fibrillation events detection from electrocardiography

OBJECTIVE

This work conducts a comparative study on the effect of neural networks of different architectures on the detection of paroxysmal atrial fibrillation (PAF) events from dynamic electrocardiography (ECG) recordings, a problem raised in the 4th China Physiological Signal Challenge 2021 (CPSC2021).

APPROACH

We proposed 3 neural network models and an auxiliary one for QRS detection to tackle the problem. A convolutional recurrent neural network (CRNN) model and a U-Net model that accepts ECG waveform input make sample-wise predictions. This regards the PAF events detection as a segmentation task. A stacked bidirectional long short-term memory (LSTM) model takes the sequence of RR intervals, which is derived from the output of the QRS detection model and makes beat-wise predictions. The QRS detection model also has a CRNN architecture, which is slightly different from the model for the AF segmentation task. Final predictions are merged by outputs from models making sample-wise predictions and making beat-wise predictions. Finally, the locations of QRS complexes are used to filter out segments (both normal and AF) shorter than 5 beats. In order to make the neural network models more sensitive to the critical sample points (onsets and offsets) of the AF events, we proposed a novel masked binary cross-entropy (MaskedBCE) loss function for training the models. This loss function is the conventional BCE loss multiplied by a mask, whose values in a neighbourhood of critical sample points are significantly larger than elsewhere.

MAIN RESULTS

Our method received a score of 1.9972 on the first part of the hidden test set of CPSC2021 and a score of 3.0907 on the second part. The average score was 2.5440, ranked 5th out of 17 teams with successful official entries.

SIGNIFICANCE

This work proposed an effective solution to the problem of the detection of PAF events from dynamic ECGs and validated the efficacy of several neural network architectures on this problem.

Device agnostic AI-based analysis of ambulatory ECG recordings

Deep Convolutional Neural Networks (DCNNs) have been shown to provide improved performance over traditional heuristic algorithms for the detection of arrhythmias from ambulatory ECG recordings. However, these DCNNs have primarily been trained and tested on device-specific databases with standardized electrode positions and uniform sampling frequencies. This work explores the possibility of training a DCNN for Atrial Fibrillation (AF) detection on a database of single‑lead ECG rhythm strips extracted from resting 12‑lead ECGs. We then test the performance of the DCNN on recordings from ambulatory ECG devices with different recording leads and sampling frequencies. We developed an extensive proprietary resting 12‑lead ECG dataset of 549,211 patients. This dataset was randomly split into a training set of 494,289 patients and a testing set of the remaining 54,922 patients. We trained a 34-layer convolutional DCNN to detect AF and other arrhythmias on this dataset. The DCNN was then validated on two Physionet databases commonly used to benchmark automated ECG algorithms (1) MIT-BIH Arrhythmia Database and (2) MIT-BIH Atrial Fibrillation Database. Validation was performed following the EC57 guidelines, with performance assessed by gross episode and duration sensitivity and positive predictive value (PPV). Finally, validation was also performed on a selection of rhythm strips from an ambulatory ECG patch that a committee of board-certified cardiologists annotated. On MIT-BIH, The DCNN achieved a sensitivity of 100% and 84% PPV in detecting episodes of AF. and 100% sensitivity and 94% PPV in quantifying AF episode duration. On AFDB, The DCNN achieved a sensitivity of 94% and PPV of 98% in detecting episodes of AF, and 98% sensitivity and 100% PPV in quantifying AF episode duration. On the patch database, the DCNN demonstrated performance that was closely comparable to that of a cardiologist. The results indicate that DCNN models can learn features that generalize between resting 12‑lead and ambulatory ECG recordings, allowing DCNNs to be device agnostic for detecting arrhythmias from single‑lead ECG recordings and enabling a range of clinical applications.

Atrioventricular Synchronization for Detection of Atrial Fibrillation and Flutter in One to Twelve ECG Leads Using a Dense Neural Network Classifier

This study investigates the use of atrioventricular (AV) synchronization as an important diagnostic criterion for atrial fibrillation and flutter (AF) using one to twelve ECG leads. Heart rate, lead-specific AV conduction time, and P-/f-wave amplitude were evaluated by three representative ECG metrics (mean value, standard deviation), namely RR-interval (RRi-mean, RRi-std), PQ-interval (PQi-mean, PQI-std), and PQ-amplitude (PQa-mean, PQa-std), in 71,545 standard 12-lead ECG records from the six largest PhysioNet CinC Challenge 2021 databases. Two rhythm classes were considered (AF, non-AF), randomly assigning records into training (70%), validation (20%), and test (10%) datasets. In a grid search of 19, 55, and 83 dense neural network (DenseNet) architectures and five independent training runs, we optimized models for one-lead, six-lead (chest or limb), and twelve-lead input features. Lead-set performance and SHapley Additive exPlanations (SHAP) input feature importance were evaluated on the test set. Optimal DenseNet architectures with the number of neurons in sequential [1st, 2nd, 3rd] hidden layers were assessed for sensitivity and specificity: DenseNet [16,16,0] with primary leads (I or II) had 87.9-88.3 and 90.5-91.5%; DenseNet [32,32,32] with six limb leads had 90.7 and 94.2%; DenseNet [32,32,4] with six chest leads had 92.1 and 93.2%; and DenseNet [128,8,8] with all 12 leads had 91.8 and 95.8%, indicating sensitivity and specificity values, respectively. Mean SHAP values on the entire test set highlighted the importance of RRi-mean (100%), RR-std (84%), and atrial synchronization (40-60%) for the PQa-mean (aVR, I), PQi-std (V2, aVF, II), and PQi-mean (aVL, aVR). Our focus on finding the strongest AV synchronization predictors of AF in 12-lead ECGs would lead to a comprehensive understanding of the decision-making process in advanced neural network classifiers. DenseNet self-learned to rely on a few ECG behavioral characteristics: first, characteristics usually associated with AF conduction such as rapid heart rate, enhanced heart rate variability, and large PQ-interval deviation in V2 and inferior leads (aVF, II); second, characteristics related to a typical P-wave pattern in sinus rhythm, which is best distinguished from AF by the earliest negative P-peak deflection of the right atrium in the lead (aVR) and late positive left atrial deflection in lateral leads (I, aVL). Our results on lead-selection and feature-selection practices for AF detection should be considered for one- to twelve-lead ECG signal processing settings, particularly those measuring heart rate, AV conduction times, and P-/f-wave amplitudes. Performances are limited to the AF diagnostic potential of these three metrics. SHAP value importance can be used in combination with a human expert's ECG interpretation to change the focus from a broad observation of 12-lead ECG morphology to focusing on the few AV synchronization findings strongly predictive of AF or non-AF arrhythmias. Our results are representative of AV synchronization findings across a broad taxonomy of cardiac arrhythmias in large 12-lead ECG databases.

Accurate detection of atrial fibrillation events with R-R intervals from ECG signals

Atrial fibrillation (AF) is a typical category of arrhythmia. Clinical diagnosis of AF is based on the detection of abnormal R-R intervals (RRIs) with an electrocardiogram (ECG). Previous studies considered this detection problem as a classification problem and focused on extracting a number of features. In this study we demonstrate that instead of using any specific numerical characteristic as the input feature, the probability density of RRIs from ECG conserves comprehensive statistical information; hence, is a natural and efficient input feature for AF detection. Incorporated with a support vector machine as the classifier, results on the MIT-BIH database indicates that the proposed method is a simple and accurate approach for AF detection in terms of accuracy, sensitivity, and specificity.

Automatic Atrial Fibrillation Arrhythmia Detection Using Univariate and Multivariate Data

Atrial fibrillation (AF) is still a major cause of disease morbidity and mortality, making its early diagnosis desirable and urging researchers to develop efficient methods devoted to automatic AF detection. Till now, the analysis of Holter-ECG recordings remains the gold-standard technique to screen AF. This is usually achieved by studying either RR interval time series analysis, P-wave detection or combinations of both morphological characteristics. After extraction and selection of meaningful features, each of the AF detection methods might be conducted through univariate and multivariate data analysis. Many of these automatic techniques have been proposed over the last years. This work presents an overview of research studies of AF detection based on RR interval time series. The aim of this paper is to provide the scientific community and newcomers to the field of AF screening with a resource that presents introductory concepts, clinical features, and a literature review that describes the techniques that are mostly followed when RR interval time series are used for accurate detection of AF.

Classification of De novo post-operative and persistent atrial fibrillation using multi-channel ECG recordings

Atrial fibrillation (AF) is the most sustained arrhythmia in the heart and also the most common complication developed after cardiac surgery. Due to its progressive nature, timely detection of AF is important. Currently, physicians use a surface electrocardiogram (ECG) for AF diagnosis. However, when the patient develops AF, its various development stages are not distinguishable for cardiologists based on visual inspection of the surface ECG signals. Therefore, severity detection of AF could start from differentiating between short-lasting AF and long-lasting AF. Here, de novo post-operative AF (POAF) is a good model for short-lasting AF while long-lasting AF can be represented by persistent AF. Therefore, we address in this paper a binary severity detection of AF for two specific types of AF. We focus on the differentiation of these two types as de novo POAF is the first time that a patient develops AF. Hence, comparing its development to a more severe stage of AF (e.g., persistent AF) could be beneficial in unveiling the electrical changes in the atrium. To the best of our knowledge, this is the first paper that aims to differentiate these different AF stages. We propose a method that consists of three sets of discriminative features based on fundamentally different aspects of the multi-channel ECG data, namely based on the analysis of RR intervals, a greyscale image representation of the vectorcardiogram, and the frequency domain representation of the ECG. Due to the nature of AF, these features are able to capture both morphological and rhythmic changes in the ECGs. Our classification system consists of a random forest classifier, after a feature selection stage using the ReliefF method. The detection efficiency is tested on 151 patients using 5-fold cross-validation. We achieved 89.07% accuracy in the classification of de novo POAF and persistent AF. The results show that the features are discriminative to reveal the severity of AF. Moreover, inspection of the most important features sheds light on the different characteristics of de novo post-operative and persistent AF.

Moving average and standard deviation thresholding (MAST): a novel algorithm for accurate R-wave detection in the murine electrocardiogram

Advances in implantable radio-telemetry or diverse biologging devices capable of acquiring high-resolution ambulatory electrocardiogram (ECG) or heart rate recordings facilitate comparative physiological investigations by enabling detailed analysis of cardiopulmonary phenotypes and responses in vivo. Two priorities guiding the meaningful adoption of such technologies are: (1) automation, to streamline and standardize large dataset analysis, and (2) flexibility in quality-control. The latter is especially relevant when considering the tendency of some fully automated software solutions to significantly underestimate heart rate when raw signals contain high-amplitude noise. We present herein moving average and standard deviation thresholding (MAST), a novel, open-access algorithm developed to perform automated, accurate, and noise-robust single-channel R-wave detection from ECG obtained in chronically instrumented mice. MAST additionally and automatically excludes and annotates segments where R-wave detection is not possible due to artefact levels exceeding signal levels. Customizable settings (e.g. window width of moving average) allow for MAST to be scaled for use in non-murine species. Two expert reviewers compared MAST's performance (true/false positive and false negative detections) with that of a commercial ECG analysis program. Both approaches were applied blindly to the same random selection of 270 3-min ECG recordings from a dataset containing varying amounts of signal artefact. MAST exhibited roughly one quarter the error rate of the commercial software and accurately detected R-waves with greater consistency and virtually no false positives (sensitivity, Se: 98.48% ± 4.32% vs. 94.59% ± 17.52%, positive predictivity, +P: 99.99% ± 0.06% vs. 99.57% ± 3.91%, P < 0.001 and P = 0.0274 respectively, Wilcoxon signed rank; values are mean ± SD). Our novel, open-access approach for automated single-channel R-wave detection enables investigators to study murine heart rate indices with greater accuracy and less effort. It also provides a foundational code for translation to other mammals, ectothermic vertebrates, and birds.

Considerations on Performance Evaluation of Atrial Fibrillation Detectors

OBJECTIVE

A large number of atrial fibrillation (AF) detectors have been published in recent years, signifying that the comparison of detector performance plays a central role, though not always consistent. The aim of this study is to shed needed light on aspects crucial to the evaluation of detection performance.

METHODS

Three types of AF detector, using either information on rhythm, rhythm and morphology, or segments of ECG samples, are implemented and studied on both real and simulated ECG signals. The properties of different performance measures are investigated, for example, in relation to dataset imbalance.

RESULTS

The results show that performance can differ considerably depending on the way detector output is compared to database annotations, i.e., beat-to-beat, segment-to-segment, or episode-to-episode comparison. Moreover, depending on the type of detector, the results substantiate that physiological and technical factors, e.g., changes in ECG morphology, rate of atrial premature beats, and noise level, can have a considerable influence on performance.

CONCLUSION

The present study demonstrates overall strengths and weaknesses of different types of detector, highlights challenges in AF detection, and proposes five recommendations on how to handle data and characterize performance.

Global hybrid multi-scale convolutional network for accurate and robust detection of atrial fibrillation using single-lead ECG recordings

BACKGROUND AND OBJECTIVE

Atrial fibrillation (AF) is the most common persistent cardiac arrhythmia in clinical practice, and its accurate screening is of great significance to avoid cardiovascular diseases (CVDs). Electrocardiogram (ECG) is considered to be the most commonly used technique for detecting AF abnormalities. However, previous ECG-based deep learning algorithms did not take into account the complementary nature of inter-layer information, which may lead to insufficient AF screening. This study reports the first attempt to use hybrid multi-scale information in a global space for accurate and robust AF detection.

METHODS

We propose a novel deep learning classification method, namely, global hybrid multi-scale convolutional neural network (i.e., GH-MS-CNN), to implement binary classification for AF detection. Unlike previous deep learning methods in AF detection, an ingenious hybrid multi-scale convolution (HMSC) module, for the advantage of automatically aggregating different types of complementary inter-layer multi-scale features in the global space, is introduced into all dense blocks of the GH-MS-CNN model to implement sufficient feature extraction, and achieve much better overall classification performance.

RESULTS

The proposed GH-MS-CNN method has been fully validated on the CPSC 2018 database and tested on the independent PhysioNet 2017 database. The experimental results show that the global and hybrid multi-scale information has tremendous advantages over local and single-type multi-scale information in AF screening. Furthermore, the proposed GH-MS-CNN method outperforms the state-of-the-art methods and achieves the best classification performance with an accuracy of 0.9984, a precision of 0.9989, a sensitivity of 0.9965, a specificity of 0.9998 and an F1 score of 0.9954. In addition, the proposed method has achieved comparable and considerable generalization capability on the PhysioNet 2017 database.

CONCLUSIONS

The proposed GH-MS-CNN method has promising capabilities and great advantages in accurate and robust AF detection. It is assumed that this research has made significant improvements in AF screening and has great potential for long-term monitoring of wearable devices.

Application of Dense Neural Networks for Detection of Atrial Fibrillation and Ranking of Augmented ECG Feature Set

Considering the significant burden to patients and healthcare systems globally related to atrial fibrillation (AF) complications, the early AF diagnosis is of crucial importance. In the view of prominent perspectives for fast and accurate point-of-care arrhythmia detection, our study optimizes an artificial neural network (NN) classifier and ranks the importance of enhanced 137 diagnostic ECG features computed from time and frequency ECG signal representations of short single-lead strips available in 2017 Physionet/CinC Challenge database. Based on hyperparameters' grid search of densely connected NN layers, we derive the optimal topology with three layers and 128, 32, 4 neurons per layer (DenseNet-3@128-32-4), which presents maximal F1-scores for classification of Normal rhythms (0.883, 5076 strips), AF (0.825, 758 strips), Other rhythms (0.705, 2415 strips), Noise (0.618, 279 strips) and total F1 relevant to the CinC Challenge of 0.804, derived by five-fold cross-validation. DenseNet-3@128-32-4 performs equally well with 137 to 32 features and presents tolerable reduction by about 0.03 to 0.06 points for limited input sets, including 8 and 16 features, respectively. The feature reduction is linked to effective application of a comprehensive method for computation of the feature map importance based on the weights of the activated neurons through the total path from input to specific output in DenseNet. The detailed analysis of 20 top-ranked ECG features with greatest importance to the detection of each rhythm and overall of all rhythms reveals DenseNet decision-making process, noticeably corresponding to the cardiologists' diagnostic point of view.

ECG data dependency for atrial fibrillation detection based on residual networks

Atrial fibrillation (AF) is an arrhythmia that can cause blood clot and may lead to stroke and heart failure. To detect AF, deep learning-based detection algorithms have recently been developed. However, deep learning models were often trained with limited datasets and were evaluated within the same datasets, which makes their performance generally drops on the external datasets, known as data dependency. For this study, three different databases from PhysioNet were used to investigate the data dependency of deep learning-based AF detection algorithm using the residual neural network (Resnet). Resnet 18, 34, 50 and 152 model were trained with raw electrocardiogram (ECG) signal extracted from independent database. The highest accuracy was about 98–99% which is evaluation results of test dataset from the own database. On the other hand, the lowest accuracy was about 53–92% which was evaluation results of the external dataset extracted from different source. There are data dependency according to the train dataset and the test dataset. However, the data dependency decreased as a large amount of train data.

Estimation of PQ distance dispersion for atrial fibrillation detection

BACKGROUND AND OBJECTIVE

Atrial fibrillation (AF) is the most common cardiac arrhythmia in the world. It is associated with significantly increased morbidity and mortality. Diagnosis of the disease can be based on the analysis of the electrical atrial activity, on quantification of the heart rate irregularity or on a mixture of the both approaches. Since the amplitude of the atrial waves is small, their analysis can lead to false results. On the other hand, the heart rate based analysis usually leads to many unnecessary warnings. Therefore, our goal is to develop a new method for effective AF detection based on the analysis of the electrical atrial waves.

METHODS

The proposed method employs the fact that there is a lack of repeatable P waves preceding QRS complexes during AF. We apply the operation of spatio-temporal filtering (STF) to magnify and detect the prominent spatio-temporal patterns (STP) within the P waves in multi-channel ECG recordings. Later we measure their distances (PQ) to the succeeding QRS complexes, and we estimate dispersion of the obtained PQ series. For signals with normal sinus rhythm, this dispersion is usually very low, and contrary, for AF it is much raised. This allows for effective discrimination of this cardiologic disorder.

RESULTS

Tested on an ECG database consisting of AF cases, normal rhythm cases and cases with normal rhythm restored by the use of cardioversion, the method proposed allowed for AF detection with the accuracy of 98.75% on the basis of both 8-channel and 2-channel signals of 12 s length. When the signals length was decreased to 6 s, the accuracy varied in the range of 95%-97.5% depending on the number of channels and the dispersion measure applied.

CONCLUSIONS

Our approach allows for high accuracy of atrial fibrillation detection using the analysis of electrical atrial activity. The method can be applied to an early detection of the desease and can advantageously be used to decrease the number of false warnings in systems based on the analysis of the heart rate.

Detection of Brief Episodes of Atrial Fibrillation Based on Electrocardiomatrix and Convolutional Neural Network

Background: Brief episodes of atrial fibrillation (AF) may evolve into longer AF episodes increasing the chances of thrombus formation, stroke, and death. Classical methods for AF detection investigate rhythm irregularity or P-wave absence in the ECG, while deep learning approaches profit from the availability of annotated ECG databases to learn discriminatory features linked to different diagnosis. However, some deep learning approaches do not provide analysis of the features used for classification. This paper introduces a convolutional neural network (CNN) approach for automatic detection of brief AF episodes based on electrocardiomatrix-images (ECM-images) aiming to link deep learning to features with clinical meaning. Materials and Methods: The CNN is trained using two databases: the Long-Term Atrial Fibrillation and the MIT-BIH Normal Sinus Rhythm, and tested on three databases: the MIT-BIH Atrial Fibrillation, the MIT-BIH Arrhythmia, and the Monzino-AF. Detection of AF is done using a sliding window of 10 beats plus 3 s. Performance is quantified using both standard classification metrics and the EC57 standard for arrhythmia detection. Layer-wise relevance propagation analysis was applied to link the decisions made by the CNN to clinical characteristics in the ECG. Results: For all three testing databases, episode sensitivity was greater than 80.22, 89.66, and 97.45% for AF episodes shorter than 15, 30 s, and for all episodes, respectively. Conclusions: Rhythm and morphological characteristics of the electrocardiogram can be learned by a CNN from ECM-images for the detection of brief episodes of AF.

Automatic Detection of Short-Term Atrial Fibrillation Segments Based on Frequency Slice Wavelet Transform and Machine Learning Techniques

Atrial fibrillation (AF) is the most frequently encountered cardiac arrhythmia and is often associated with other cardiovascular and cerebrovascular diseases, such as ischemic heart disease, chronic heart failure, and stroke. Automatic detection of AF by analyzing electrocardiogram (ECG) signals has an important application value. Using the contaminated and actual ECG signals, it is not enough to only analyze the atrial activity of disappeared P wave and appeared F wave in the TQ segment. Moreover, the best analysis method is to combine nonlinear features analyzing ventricular activity based on the detection of R peak. In this paper, to utilize the information of the P-QRS-T waveform generated by atrial and ventricular activity, frequency slice wavelet transform (FSWT) is adopted to conduct time-frequency analysis on short-term ECG segments from the MIT-BIH Atrial Fibrillation Database. The two-dimensional time-frequency matrices are obtained. Furthermore, an average sliding window is used to convert the two-dimensional time-frequency matrices to the one-dimensional feature vectors, which are classified using five machine learning (ML) techniques. The experimental results show that the classification performance of the Gaussian-kernel support vector machine (GKSVM) based on the Bayesian optimizer is better. The accuracy of the training set and validation set are 100% and 93.4%. The accuracy, sensitivity, and specificity of the test set without training are 98.15%, 96.43%, and 100%, respectively. Compared with previous research results, our proposed FSWT-GKSVM model shows stability and robustness, and it could achieve the purpose of automatic detection of AF.

Real-Time Heart Arrhythmia Detection Using Apache Spark Structured Streaming

One of the major causes of death in the world is cardiac arrhythmias. In the field of healthcare, physicians use the patient's electrocardiogram (ECG) records to detect arrhythmias, which indicate the electrical activity of the patient's heart. The problem is that the symptoms do not always appear and the physician may be mistaken in the diagnosis. Therefore, patients need continuous monitoring through real-time ECG analysis to detect arrhythmias in a timely manner and prevent an eventual incident that threatens the patient's life. In this research, we used the Structured Streaming module built top on the open-source Apache Spark platform for the first time to implement a machine learning pipeline for real-time cardiac arrhythmias detection and evaluate the impact of using this new module on classification performance metrics and the rate of delay in arrhythmia detection. The ECG data collected from the MIT/BIH database for the detection of three class labels: normal beats, RBBB, and atrial fibrillation arrhythmias. We also developed three decision trees, random forest, and logistic regression multiclass classifiers for data classification where the random forest classifier showed better performance in classification than the other two classifiers. The results show previous results in performance metrics of the classification model and a significant decrease in pipeline runtime by using more class labels compared to previous studies.

Atrial Fibrillation Detection During Sepsis: Study on MIMIC III ICU Data

Sepsis is defined by life-threatening organ dysfunction during infection and is one of the leading causes of critical illness. During sepsis, there is high risk that new-onset of atrial fibrillation (AF) can occur, which is associated with significant morbidity and mortality. As a result, computer aided automated and reliable detection of new-onset AF during sepsis is crucial, especially for the critically ill patients in the intensive care unit (ICU). In this paper, a novel automated and robust two-step algorithm to detect AF from ICU patients using electrocardiogram (ECG) signals is presented. First, several statistical parameters including root mean square of successive differences, Shannon entropy, and sample entropy were calculated from the heart rate for the screening of possible AF segments. Next, Poincaré plot-based features along with P-wave characteristics were used to reduce false positive detection of AF, caused by the premature atrial and ventricular beats. A subset of the Medical Information Mart for Intensive Care (MIMIC) III database containing 198 subjects was used in this study. During the training and validation phases, both the simple thresholding as well as machine learning classifiers achieved very high segment-wise AF classification performance. Finally, we tested the performance of our proposed algorithm using two independent test data sets and compared the performance with two state-of-the-art methods. The algorithm achieved an overall 100% sensitivity, 98% specificity, 98.99% accuracy, 98% positive predictive value, and 100% negative predictive value on the subject-wise AF detection, thus showing the efficacy of our proposed algorithm in critically ill sepsis patients. The annotations of the data have been made publicly available for other investigators.

HAN-ECG: An interpretable atrial fibrillation detection model using hierarchical attention networks

Atrial fibrillation (AF) is one of the most prevalent cardiac arrhythmias that affects the lives of many people around the world and is associated with a five-fold increased risk of stroke and mortality. Like other problems in the healthcare domain, artificial intelligence (AI)-based models have been used to detect AF from patients' ECG signals. The cardiologist level performance in detecting this arrhythmia is often achieved by deep learning-based methods, however, they suffer from the lack of interpretability. In other words, these approaches are unable to explain the reasons behind their decisions. The lack of interpretability is a common challenge toward a wide application of machine learning (ML)-based approaches in the healthcare which limits the trust of clinicians in such methods. To address this challenge, we propose HAN-ECG, an interpretable bidirectional-recurrent-neural-network-based approach for the AF detection task. The HAN-ECG employs three attention mechanism levels to provide a multi-resolution analysis of the patterns in ECG leading to AF. The detected patterns by this hierarchical attention model facilitate the interpretation of the neural network decision process in identifying the patterns in the signal which contributed the most to the final detection. Experimental results on two AF databases demonstrate that our proposed model performs better than the existing algorithms. Visualization of these attention layers illustrates that our proposed model decides upon the important waves and heartbeats which are clinically meaningful in the detection task (e.g., absence of P-waves, and irregular R-R intervals for the AF detection task).

Computer Aided Diagnosis for atrial fibrillation based on new artificial adaptive systems

BACKGROUND AND OBJECTIVE

Atrial fibrillation (AF) is the most common cardiac arrhythmia in clinical practice, having been recognized as a true cardiovascular epidemic. In this paper, a new methodology for Computer Aided Diagnosis of AF based on a special kind of artificial adaptive systems has been developed.

METHODS

Following the extraction of data from the PhysioNet repository, a new dataset composed of the R/R distances of 73 patients was created. To avoid redundancy, the training set was created by randomly selecting 50% of the subjects from the entire sample, thus making a choice by patient and not by record. The remaining 50% of subjects were randomly split by records in testing and prediction sets. The original ECG data has been transformed according to the following four orders of abstraction: a) sequence of R/R intervals; b) composition of ECG data into a moving window; c) training of different machine learning systems to abstract the function governing the AF; d) fuzzy transformation of Machine learning estimations. In this paper, in parallel with the classic method of windowing, we propose a variant based on a system of progressive moving averages.

RESULTS

The best performing machine learning, Supervised Contractive Map (SVCm), reached an overall mean accuracy of 95%. SVCm is a new deep neural network based on a different principle than the usual descending gradient. The minimization of the error occurs by means of decomposition into contracted sine functions.

CONCLUSIONS

In this research, atrial fibrillation is considered from a completely different point of view than classical methods. It is seen as the stable process, i.e. the function, that manages the irregularity of the irregularities of the R/R intervals. The idea, therefore, is to abstract from mere physiology to investigate fibrillation as a mathematical object that handles irregularities. The attained results seem to open new perspectives for the use of potent artificial adaptive systems for the automatic detection of atrial fibrillation, with accuracy rates extremely promising for real world applications.

Comparative analysis between convolutional neural network learned and engineered features: A case study on cardiac arrhythmia detection

Background

Atrial fibrillation (AF) is one of the most common cardiovascular problems, and its asymptomatic tendency makes AF detection challenging. Machine and deep learning methods are commonly used in AF detection.

Objective

The purpose of this study was to evaluate the information provided by convolutional neural network (CNN) and random forest (RF) machine learning models for AF classification.

Methods

We manually extracted 166 time-frequency domains and linear and nonlinear features to classify single-lead electrocardiograms (ECGs) as normal, AF, other, or noisy sinus rhythms. We selected a subset of 56 robust features using a genetic algorithm that was used in the RF model. In a separate study, a 1-dimensional, 12-layer CNN was designed on the raw ECG rhythms. Four features from the output layer and 128 features from the fully connected layer of CNN were explored independently for classification. The models were trained and internally validated on 8,528 ECGs and externally validated on a hidden dataset containing 3,658 ECGs. Next,we analyzed the correlation between engineered and CNN-learned features.

Results

An RF classifier trained with 56-engineered features resulted in an F₁ score of 0.91, 0.78, and 0.72 for normal, AF, and other rhythms, respectively. However, an ensemble of support vector machine and the CNN model resulted in an F₁ score of 0.92, 0.87, and 0.80, respectively.

Conclusion

We explored various features and machine learning models to identify AF rhythms using short (9-61 seconds) single-lead ECG recordings. Our results showed that the proposed CNN model abstracted distinctive features for AF classification.

Electronics and Its Worldwide Research

The contributions of researchers at a global level in the journal Electronics in the period 2012–2020 are analyzed. The objective of this work is to establish a global vision of the issues published in the Electronic magazine and their importance, advances and developments that have been particularly relevant for subsequent research. The magazine has 15 thematic sections and a general one, with the programming of 385 special issues for 2020–2021. Using the Scopus database and bibliometric techniques, 2310 documents are obtained and distributed in 14 thematic communities. The communities that contribute to the greatest number of works are Power Electronics (20.13%), Embedded Computer Systems (13.59%) and Internet of Things and Machine Learning Systems (8.11%). A study of the publications by authors, affiliations, countries as well as the H index was undertaken. The 7561 authors analyzed are distributed in 87 countries, with China being the country of the majority (2407 authors), followed by South Korea (763 authors). The H-index of most authors (75.89%) ranges from 0 to 9, where the authors with the highest H-Index are from the United States, Denmark, Italy and India. The main publication format is the article (92.16%) and the review (5.84%). The magazine publishes topics in continuous development that will be further investigated and published in the near future in fields as varied as the transport sector, energy systems, the development of new broadband semiconductors, new modulation and control techniques, and more.

Convolutional Neural Network Based Detection of Atrial Fibrillation Combing R-R intervals and F-wave Frequency Spectrum

Atrial Fibrillation (AF) is one of the arrhythmias that is common and serious in clinic. In this study, a novel method of AF classification with a convolutional neural network (CNN) was proposed, and particularly two cardiac rhythm features of R-R intervals and F-wave frequency spectrum were combined into the CNN for a good applicability of mobile application. Over 23 patients' ten-hours of Electrocardiogram (ECG) records were collected from the MIT-BIH database, and each of which was segmented into 10s-data fragments to train the designed CNN and evaluate the performance of the proposed method. Specifically, a total of 83,461 fragments were collected, 49,952 fragments of which are the normal fragments (type-N) and the others are the AF fragments. As results, the obtained average accuracy of the proposed method combining the two proposed features is 97.3%, which is shown a relative higher accuracy comparing with either that of the detection with the feature of R-R intervals (95.7%) or that with the feature of F-wave frequency spectrum (93.9%). Additionally, the sensitivity and the specificity of the present method are both of a high level of 97.4% and 97.2%, respectively. In conclusion, the CNN based approach by combining the R-R interval series and the F-wave frequency spectrum would be effectively to improve the performance of AF detection. Moreover, the proposed classification of AF with 10s-data fragments also could be potentially useful for a wearable real-time monitoring application for a pre-hospital screening of AF.

SS-SWT and SI-CNN: An Atrial Fibrillation Detection Framework for Time-Frequency ECG Signal

Atrial fibrillation is the most common arrhythmia and is associated with high morbidity and mortality from stroke, heart failure, myocardial infarction, and cerebral thrombosis. Effective and rapid detection of atrial fibrillation is critical to reducing morbidity and mortality in patients. Screening atrial fibrillation quickly and efficiently remains a challenging task. In this paper, we propose SS-SWT and SI-CNN: an atrial fibrillation detection framework for the time-frequency ECG signal. First, specific-scale stationary wavelet transform (SS-SWT) is used to decompose a 5-s ECG signal into 8 scales. We select specific scales of coefficients as valid time-frequency features and abandon the other coefficients. The selected coefficients are fed to the scale-independent convolutional neural network (SI-CNN) as a two-dimensional (2D) matrix. In SI-CNN, a convolution kernel specifically for the time-frequency characteristics of ECG signals is designed. During the convolution process, the independence between each scale of coefficient is preserved, and the time domain and the frequency domain characteristics of the ECG signal are effectively extracted, and finally the atrial fibrillation signal is quickly and accurately identified. In this study, experiments are performed using the MIT-BIH AFDB data in 5-s data segments. We achieve 99.03% sensitivity, 99.35% specificity, and 99.23% overall accuracy. The SS-SWT and SI-CNN we propose simplify the feature extraction step, effectively extracts the features of ECG, and reduces the feature redundancy that may be caused by wavelet transform. The results shows that the method can effectively detect atrial fibrillation signals and has potential in clinical application.

Detection of Atrial Fibrillation from Single Lead ECG Signal Using Multirate Cosine Filter Bank and Deep Neural Network

Atrial fibrillation (AF) is a cardiac arrhythmia which is characterized based on the irregsular beating of atria, resulting in, the abnormal atrial patterns that are observed in the electrocardiogram (ECG) signal. The early detection of this pathology is very helpful for minimizing the chances of stroke, other heart-related disorders, and coronary artery diseases. This paper proposes a novel method for the detection of AF pathology based on the analysis of the ECG signal. The method adopts a multi-rate cosine filter bank architecture for the evaluation of coefficients from the ECG signal at different subbands, in turn, the Fractional norm (FN) feature is evaluated from the extracted coefficients at each subband. Then, the AF detection is carried out using a deep learning approach known as the Hierarchical Extreme Learning Machine (H-ELM) from the FN features. The proposed method is evaluated by considering normal and AF pathological ECG signals from public databases. The experimental results reveal that the proposed multi-rate cosine filter bank based on FN features is effective for the detection of AF pathology with an accuracy, sensitivity and specificity values of 99.40%, 98.77%, and 100%, respectively. The performance of the proposed diagnostic features of the ECG signal is compared with other existing features for the detection of AF. The low-frequency subband FN features found to be more significant with a difference of the mean values as 0.69 between normal and AF classes.

A Review of Atrial Fibrillation Detection Methods as a Service

Atrial Fibrillation (AF) is a common heart arrhythmia that often goes undetected, and even if it is detected, managing the condition may be challenging. In this paper, we review how the RR interval and Electrocardiogram (ECG) signals, incorporated into a monitoring system, can be useful to track AF events. Were such an automated system to be implemented, it could be used to help manage AF and thereby reduce patient morbidity and mortality. The main impetus behind the idea of developing a service is that a greater data volume analyzed can lead to better patient outcomes. Based on the literature review, which we present herein, we introduce the methods that can be used to detect AF efficiently and automatically via the RR interval and ECG signals. A cardiovascular disease monitoring service that incorporates one or multiple of these detection methods could extend event observation to all times, and could therefore become useful to establish any AF occurrence. The development of an automated and efficient method that monitors AF in real time would likely become a key component for meeting public health goals regarding the reduction of fatalities caused by the disease. Yet, at present, significant technological and regulatory obstacles remain, which prevent the development of any proposed system. Establishment of the scientific foundation for monitoring is important to provide effective service to patients and healthcare professionals.

Detection of Atrial Fibrillation Episodes in Long-Term Heart Rhythm Signals Using a Support Vector Machine

Atrial fibrillation (AF) is a serious heart arrhythmia leading to a significant increase of the risk for occurrence of ischemic stroke. Clinically, the AF episode is recognized in an electrocardiogram. However, detection of asymptomatic AF, which requires a long-term monitoring, is more efficient when based on irregularity of beat-to-beat intervals estimated by the heart rate (HR) features. Automated classification of heartbeats into AF and non-AF by means of the Lagrangian Support Vector Machine has been proposed. The classifier input vector consisted of sixteen features, including four coefficients very sensitive to beat-to-beat heart changes, taken from the fetal heart rate analysis in perinatal medicine. Effectiveness of the proposed classifier has been verified on the MIT-BIH Atrial Fibrillation Database. Designing of the LSVM classifier using very large number of feature vectors requires extreme computational efforts. Therefore, an original approach has been proposed to determine a training set of the smallest possible size that still would guarantee a high quality of AF detection. It enables to obtain satisfactory results using only 1.39% of all heartbeats as the training data. Post-processing stage based on aggregation of classified heartbeats into AF episodes has been applied to provide more reliable information on patient risk. Results obtained during the testing phase showed the sensitivity of 98.94%, positive predictive value of 98.39%, and classification accuracy of 98.86%.

Atrial Fibrillation Detection in Short Single Lead ECG Recordings Using Wavelet Transform and Artificial Neural Networks

Atrial fibrillation (AF) is a common health issue, not only in developed countries but also in developing ones. AF can lead to strokes, heart failures, and even death if it is not diagnosed and treated on time, therefore automatic detection of AF is an urgent need, particularly using Internet- connected devices that can alert healthcare services. Detection of AF typically involves the analysis of electrocardiogram (ECG) recordings, where P-waves that characterize the atrial activity are substituted with f-waves of variable amplitude and duration. In this paper, we used the discrete wavelet transform to decompose the ECG signal into detail and approximation coefficients with different time-frequency resolutions. Features extracted from ECG signals, RR interval time series and detail and approximation coefficients were used as inputs to an artificial neural network trained to identify four classes of heart rhythms: normal sinus rhythm (NSR), AF, other rhythms (OR) and noisy signals (NS). By performing a Monte Carlo 10- fold cross-validation of 10 iterations approach, average micro F₁ scores of 83.64%, 61.61%, 56.88% and 53.88% to classify NSR, AF, OR and NS respectively, and average macro F₁ of 64.00% were obtained on the publicly available training set of PhysioNet/Computing in Cardiology Challenge 2017. In addition, in a one-vs.-the-rest strategy, i.e., AF-vs-the-rest, averages sensitivity and specificity of 95.70% and 72.39% respectively were achieved to classify AF recordings.

ECGNET: Learning where to attend for detection of atrial fibrillation with deep visual attention

The complexity of the patterns associated with atrial fibrillation (AF) and the high level of noise affecting these patterns have significantly limited the application of current signal processing and shallow machine learning approaches to accurately detect this condition. Deep neural networks have shown to be very powerful to learn the non-linear patterns in various problems such as computer vision tasks. While deep learning approaches have been utilized to learn complex patterns related to the presence of AF in electrocardiogram (ECG) signals, they can considerably benefit from knowing which parts of the signal is more important to focus on during learning. In this paper, we introduce a two-channel deep neural network to more accurately detect the presence of AF in the ECG signals. The first channel takes in an ECG signal and automatically learns where to attend for detection of AF. The second channel simultaneously takes in the same ECG signal to consider all features of the entire signal. Besides improving detection accuracy, this model can guide the physicians via visualization that what parts of the given ECG signal are important to attend while trying to detect atrial fibrillation. The experimental results confirm that the proposed model significantly improves the performance of AF detection on well-known MIT-BIH AF database with 5-s ECG segments (achieved a sensitivity of 99.53%, specificity of 99.26% and accuracy of 99.40%).

Automatic Prediction of Atrial Fibrillation Based on Convolutional Neural Network Using a Short-term Normal Electrocardiogram Signal

BACKGROUND

In this study, we propose a method for automatically predicting atrial fibrillation (AF) based on convolutional neural network (CNN) using a short-term normal electrocardiogram (ECG) signal.

METHODS

We designed a CNN model and optimized it by dropout and normalization. One-dimensional convolution, max-pooling, and fully-connected multiple perceptron were used to analyze the short-term normal ECG. The ECG signal was preprocessed and segmented to train and evaluate the proposed CNN model. The training and test sets consisted of the two AF and one normal dataset from the MIT-BIH database.

RESULTS

The proposed CNN model for the automatic prediction of AF achieved a high performance with a sensitivity of 98.6%, a specificity of 98.7%, and an accuracy of 98.7%.

CONCLUSION

The results show the possibility of automatically predicting AF based on the CNN model using a short-term normal ECG signal. The proposed CNN model for the automatic prediction of AF can be a helpful tool for the early diagnosis of AF in healthcare fields.

Atrial Fibrillation Detection by the Combination of Recurrence Complex Network and Convolution Neural Network

In this paper, R wave peak interval independent atrial fibrillation detection algorithm is proposed based on the analysis of the synchronization feature of the electrocardiogram signal by a deep neural network. Firstly, the synchronization feature of each heartbeat of the electrocardiogram signal is constructed by a Recurrence Complex Network. Then, a convolution neural network is used to detect atrial fibrillation by analyzing the eigenvalues of the Recurrence Complex Network. Finally, a voting algorithm is developed to improve the performance of the beat-wise atrial fibrillation detection. The MIT-BIH atrial fibrillation database is used to evaluate the performance of the proposed method. Experimental results show that the sensitivity, specificity, and accuracy of the algorithm can achieve 94.28%, 94.91%, and 94.59%, respectively. Remarkably, the proposed method was more effective than the traditional algorithms to the problem of individual variation in the atrial fibrillation detection.

A New Entropy-Based Atrial Fibrillation Detection Method for Scanning Wearable ECG Recordings

Entropy-based atrial fibrillation (AF) detectors have been applied for short-term electrocardiogram (ECG) analysis. However, existing methods suffer from several limitations. To enhance the performance of entropy-based AF detectors, we have developed a new entropy measure, named Entropy_AF, which includes the following improvements: (1) use of a ranged function rather than the Chebyshev function to define vector distance, (2) use of a fuzzy function to determine vector similarity, (3) replacement of the probability estimation with density estimation for entropy calculation, (4) use of a flexible distance threshold parameter, and (5) use of adjusted entropy results for the heart rate effect. Entropy_AF was trained using the MIT-BIH Atrial Fibrillation (AF) database, and tested on the clinical wearable long-term AF recordings. Three previous entropy-based AF detectors were used for comparison: sample entropy (SampEn), fuzzy measure entropy (FuzzyMEn) and coefficient of sample entropy (COSEn). For classifying AF and non-AF rhythms in the MIT-BIH AF database, Entropy_AF achieved the highest area under receiver operating characteristic curve (AUC) values of 98.15% when using a 30-beat time window, which was higher than COSEn with AUC of 91.86%. SampEn and FuzzyMEn resulted in much lower AUCs of 74.68% and 79.24% respectively. For classifying AF and non-AF rhythms in the clinical wearable AF database, Entropy_AF also generated the largest values of Youden index (77.94%), sensitivity (92.77%), specificity (85.17%), accuracy (87.10%), positive predictivity (68.09%) and negative predictivity (97.18%). COSEn had the second-best accuracy of 78.63%, followed by an accuracy of 65.08% in FuzzyMEn and an accuracy of 59.91% in SampEn. The new proposed Entropy_AF also generated highest classification accuracy when using a 12-beat time window. In addition, the results from time cost analysis verified the efficiency of the new Entropy_AF. This study showed the better discrimination ability for identifying AF when using Entropy_AF method, indicating that it would be useful for the practical clinical wearable AF scanning.

Detection of Atrial Fibrillation from RR Intervals and PQRST Morphology using a Neural Network Ensemble

Early detection and discrimination of cardiac arrhythmia, atrial fibrillation (AF) in particular, is essential for timely intervention to improve patient outcomes. In this work, an algorithm was developed to classify ECG records as normal, AF, other arrhythmia, or too noisy to classify. This algorithm, which was an entry for the PhysioNet Computing in Cardiology Challenge 2017 (the Challenge), is described. Artifact masking and QRS detection were applied to lead-I equivalent ECG records and 17 features were extracted which captured the irregularity of the RR intervals, the PQRST morphology, and artifact/noise. An ensemble of ten neural networks (NN) was trained on the features from a training set of 5,970 records. A final classification was taken by majority vote over the 10 classifiers. The trained NN models were validated on a further 2,558 ECG records and then tested on a blind out-of-sample test set of 3,658 records. A mean $F_{1}$ score across the four classes of 0.78 for the training/validation sets and 0.80 for the testing set was achieved. A higher $F_{1}$ score for the testing set indicates that overtraining did not occur, unlike most entries to the Challenge (winner mean $F_{1}$ score of 0.89 for training/validation set, and 0.83 for testing set). Performance of the Challenge winner was not ideal and there is evidence of overtraining, indicating the difficulty of classifying AF from single-lead ECG. The features and method described here performed comparably and overtraining did not occur (high likelihood of generalization) indicating a good starting point for future work.

ECG signal classification for the detection of cardiac arrhythmias using a convolutional recurrent neural network

OBJECTIVE

The electrocardiogram (ECG) provides an effective, non-invasive approach for clinical diagnosis in patients with cardiac diseases such as atrial fibrillation (AF). AF is the most common cardiac rhythm disturbance and affects ~2% of the general population in industrialized countries. Automatic AF detection in clinics remains a challenging task due to the high inter-patient variability of ECGs, and unsatisfactory existing approaches for AF diagnosis (e.g. atrial or ventricular activity-based analyses).

APPROACH

We have developed RhythmNet, a 21-layer 1D convolutional recurrent neural network, trained using 8528 single-lead ECG recordings from the 2017 PhysioNet/Computing in Cardiology (CinC) Challenge, to classify ECGs of different rhythms including AF automatically. Our RhythmNet architecture contained 16 convolutions to extract features directly from raw ECG waveforms, followed by three recurrent layers to process ECGs of varying lengths and to detect arrhythmia events in long recordings. Large 15  ×  1 convolutional filters were used to effectively learn the detailed variations of the signal within small time-frames such as the P-waves and QRS complexes. We employed residual connections throughout RhythmNet, along with batch-normalization and rectified linear activation units to improve convergence during training.

MAIN RESULTS

We evaluated our algorithm on 3658 testing data and obtained an F 1 accuracy of 82% for classifying sinus rhythm, AF, and other arrhythmias. RhythmNet was also ranked 5th in the 2017 CinC Challenge.

SIGNIFICANCE

Potentially, our approach could aid AF diagnosis in clinics and be used for patient self-monitoring to improve the early detection and effective treatment of AF.