Expert-enhanced machine learning for cardiac arrhythmia classification.

We propose a new method for the classification task of distinguishing atrial fibrillation (AFib) from regular atrial tachycardias including atrial flutter (AFlu) based on a surface electrocardiogram (ECG). Recently, many approaches for an automatic classification of cardiac arrhythmia were proposed and to our knowledge none of them can distinguish between these two. We discuss reasons why deep learning may not yield satisfactory results for this task. We generate new and clinically interpretable features using mathematical optimization for subsequent use within a machine learning (ML) model. These features are generated from the same input data by solving an additional regression problem with complicated combinatorial substructures. The resultant can be seen as a novel machine learning model that incorporates expert knowledge on the pathophysiology of atrial flutter. Our approach achieves an unprecedented accuracy of 82.84% and an area under the receiver operating characteristic (ROC) curve of 0.9, which classifies as "excellent" according to the classification indicator of diagnostic tests. One additional advantage of our approach is the inherent interpretability of the classification results. Our features give insight into a possibly occurring multilevel atrioventricular blocking mechanism, which may improve treatment decisions beyond the classification itself. Our research ideally complements existing textbook cardiac arrhythmia classification methods, which cannot provide a classification for the important case of AFib↔AFlu. The main contribution is the successful use of a novel mathematical model for multilevel atrioventricular block and optimization-driven inverse simulation to enhance machine learning for classification of the arguably most difficult cases in cardiac arrhythmia. A tailored Branch-and-Bound algorithm was implemented for the domain knowledge part, while standard algorithms such as Adam could be used for training.

Convolutional neural network for classification of eight types of arrhythmia using 2D time-frequency feature map from standard 12-lead electrocardiogram.

Electrocardiograms (ECGs) are widely used for diagnosing cardiac arrhythmia based on the deformation of signal shapes due to changes in various heart diseases. However, these abnormal signs may not be observed in some 12 ECG channels, depending on the location, the heart shape, and the type of cardiac arrhythmia. Therefore, it is necessary to closely and comprehensively observe ECG records acquired from 12 channel electrodes to diagnose cardiac arrhythmias accurately. In this study, we proposed a clustering algorithm that can classify persistent cardiac arrhythmia as well as episodic cardiac arrhythmias using the standard 12-lead ECG records and the 2D CNN model using the time-frequency feature maps to classify the eight types of arrhythmias and normal sinus rhythm. The standard 12-lead ECG records were provided by China Physiological Signal Challenge 2018 and consisted of 6877 patients. The proposed algorithm showed high performance in classifying persistent cardiac arrhythmias; however, its accuracy was somewhat low in classifying episodic arrhythmias. If our proposed model is trained and verified using more clinical data, we believe it can be used as an auxiliary device for diagnosing cardiac arrhythmias.

Interpretation of Electrocardiogram Heartbeat by CNN and GRU.

The diagnosis of electrocardiogram (ECG) is extremely onerous and inefficient, so it is necessary to use a computer-aided diagnosis of ECG signals. However, it is still a challenging problem to design high-accuracy ECG algorithms suitable for the medical field. In this paper, a classification method is proposed to classify ECG signals. Firstly, wavelet transform is used to denoise the original data, and data enhancement technology is used to overcome the problem of an unbalanced dataset. Secondly, an integrated convolutional neural network (CNN) and gated recurrent unit (GRU) classifier is proposed. The proposed network consists of a convolution layer, followed by 6 local feature extraction modules (LFEM), a GRU, and a Dense layer and a Softmax layer. Finally, the processed data were input into the CNN-GRU network into five categories: nonectopic beats, supraventricular ectopic beats, ventricular ectopic beats, fusion beats, and unknown beats. The MIT-BIH arrhythmia database was used to evaluate the approach, and the average sensitivity, accuracy, and F1-score of the network for 5 types of ECG were 99.33%, 99.61%, and 99.42%. The evaluation criteria of the proposed method are superior to other state-of-the-art methods, and this model can be applied to wearable devices to achieve high-precision monitoring of ECG.

ECG Heartbeat Classification Based on an Improved ResNet-18 Model.

Based on a convolutional neural network (CNN) approach, this article proposes an improved ResNet-18 model for heartbeat classification of electrocardiogram (ECG) signals through appropriate model training and parameter adjustment. Due to the unique residual structure of the model, the utilized CNN layered structure can be deepened in order to achieve better classification performance. The results of applying the proposed model to the MIT-BIH arrhythmia database demonstrate that the model achieves higher accuracy (96.50%) compared to other state-of-the-art classification models, while specifically for the ventricular ectopic heartbeat class, its sensitivity is 93.83% and the precision is 97.44%.

Mechanical dyssynchrony: How do we measure it, what it means, and what we can do about it.

Left ventricular mechanical dyssynchrony (LVMD) is defined by a difference in the timing of mechanical contraction or relaxation between different segments of the left ventricle (LV). Mechanical dyssynchrony is distinct from electrical dyssynchrony as measured by QRS duration and has been of increasing interest due to its association with worse prognosis and potential role in patient selection for cardiac resynchronization therapy (CRT). Although echocardiography is the most used modality to assess LVMD, some limitations apply to this modality. Compared to echo-based modalities, nuclear imaging by gated single-photon emission computed tomography (GSPECT) myocardial perfusion imaging (MPI) has clear advantages in evaluating systolic and diastolic LVMD. GSPECT MPI can determine systolic and diastolic mechanical dyssynchrony by the variability in the timing in which different LV segments contract or relax, which has prognostic impact in patients with coronary artery disease and heart failure. As such, by targeting mechanical dyssynchrony instead of electrical dyssynchrony, GSPECT MPI can potentially improve patient selection for CRT. So far, few studies have investigated the role of diastolic dyssynchrony, but recent evidence seems to suggest high prevalence and more prognostic impact than previously recognized. In the present review, we provide an oversight of mechanical dyssynchrony.

Usefulness of Machine Learning-Based Detection and Classification of Cardiac Arrhythmias With 12-Lead Electrocardiograms.

BACKGROUND: Deep-learning algorithms to annotate electrocardiograms (ECGs) and classify different types of cardiac arrhythmias with the use of a single-lead ECG input data set have been developed. It remains to be determined whether these algorithms can be generalized to 12-lead ECG-based rhythm classification. METHODS: We used a long short-term memory (LSTM) model to detect 12 heart rhythm classes with the use of 65,932 digital 12-lead ECG signals from 38,899 patients, using annotations obtained by consensus of 3 board-certified electrophysiologists as the criterion standard. RESULTS: The accuracy of the LSTM model for the classification of each of the 12 heart rhythms was ≥ 0.982 (range 0.982-1.0), with an area under the receiver operating characteristic curve of ≥ 0.987 (range 0.987-1.0). The precision and recall ranged from 0.692 to 1 and from 0.625 to 1, respectively, with an F1 score of ≥ 0.777 (range 0.777-1.0). The accuracy of the model (0.90) was superior to the mean accuracies of internists (0.55), emergency physicians (0.73), and cardiologists (0.83). CONCLUSIONS: We demonstrated the feasibility and effectiveness of the deep-learning LSTM model for interpreting 12 common heart rhythms according to 12-lead ECG signals. The findings may have clinical relevance for the early diagnosis of cardiac rhythm disorders.

Wavelet Scattering Transform for ECG Beat Classification.

An electrocardiogram (ECG) records the electrical activity of the heart; it contains rich pathological information on cardiovascular diseases, such as arrhythmia. However, it is difficult to visually analyze ECG signals due to their complexity and nonlinearity. The wavelet scattering transform can generate translation-invariant and deformation-stable representations of ECG signals through cascades of wavelet convolutions with nonlinear modulus and averaging operators. We proposed a novel approach using wavelet scattering transform to automatically classify four categories of arrhythmia ECG heartbeats, namely, nonectopic (N), supraventricular ectopic (S), ventricular ectopic (V), and fusion (F) beats. In this study, the wavelet scattering transform extracted 8 time windows from each ECG heartbeat. Two dimensionality reduction methods, principal component analysis (PCA) and time window selection, were applied on the 8 time windows. These processed features were fed to the neural network (NN), probabilistic neural network (PNN), and k-nearest neighbour (KNN) classifiers for classification. The 4th time window in combination with KNN (k = 4) has achieved the optimal performance with an averaged accuracy, positive predictive value, sensitivity, and specificity of 99.3%, 99.6%, 99.5%, and 98.8%, respectively, using tenfold cross-validation. Thus, our proposed model is capable of highly accurate arrhythmia classification and will provide assistance to physicians in ECG interpretation.

Electrocardiographic features of patients with COVID-19 pneumonia.

BACKGROUND: . The electrocardiographic (ECG) changes which may occur during hospitalization for COVID-19 have not yet been comprehensively assessed. PATIENTS AND METHODS: . We examined 50 patients admitted to hospital with proven COVID-19 pneumonia. At entry, all patients underwent a detailed clinical examination, 12-lead ECG, laboratory tests and arterial blood gas test. ECG was also recorded at discharge and in case of worsening clinical conditions. RESULTS: . Mean age of patients was 64 years and 72% were men. At baseline, 30% of patients had ST-T abnormalities, and 33% had left ventricular hypertrophy. During hospitalization, 26% of patients developed new ECG abnormalities which included atrial fibrillation, ST-T changes, tachy-brady syndrome, and changes consistent with acute pericarditis. One patient was transferred to intensive care unit for massive pulmonary embolism with right bundle branch block, and another for non-ST segment elevation myocardial infarction. Patients free of ECG changes during hospitalization were more likely to be treated with antiretrovirals (68% vs 15%, p = 0.001) and hydroxychloroquine (89% vs 62%, p = 0.026) versus those who developed ECG abnormalities after admission. Most measurable ECG features at discharge did not show significant changes from baseline (all p>0.05) except for a slightly decrease in Cornell voltages (13±6 vs 11±5 mm; p = 0.0001) and a modest increase in the PR interval. The majority (54%) of patients with ECG abnormalities had 2 prior consecutive negative nasopharyngeal swabs. ECG abnormalities were first detected after an average of about 30 days from symptoms' onset (range 12-51 days). CONCLUSIONS: . ECG abnormalities during hospitalization for COVID-19 pneumonia reflect a wide spectrum of cardiovascular complications, exhibit a late onset, do not progress in parallel with pulmonary abnormalities and may occur after negative nasopharyngeal swabs.

A Metaheuristic Optimization Approach for Parameter Estimation in Arrhythmia Classification from Unbalanced Data.

The electrocardiogram records the heart's electrical activity and generates a significant amount of data. The analysis of these data helps us to detect diseases and disorders via heart bio-signal abnormality classification. In unbalanced-data contexts, where the classes are not equally represented, the optimization and configuration of the classification models are highly complex, reflecting on the use of computational resources. Moreover, the performance of electrocardiogram classification depends on the approach and parameter estimation to generate the model with high accuracy, sensitivity, and precision. Previous works have proposed hybrid approaches and only a few implemented parameter optimization. Instead, they generally applied an empirical tuning of parameters at a data level or an algorithm level. Hence, a scheme, including metrics of sensitivity in a higher precision and accuracy scale, deserves special attention. In this article, a metaheuristic optimization approach for parameter estimations in arrhythmia classification from unbalanced data is presented. We selected an unbalanced subset of those databases to classify eight types of arrhythmia. It is important to highlight that we combined undersampling based on the clustering method (data level) and feature selection method (algorithmic level) to tackle the unbalanced class problem. To explore parameter estimation and improve the classification for our model, we compared two metaheuristic approaches based on differential evolution and particle swarm optimization. The final results showed an accuracy of 99.95%, a F1 score of 99.88%, a sensitivity of 99.87%, a precision of 99.89%, and a specificity of 99.99%, which are high, even in the presence of unbalanced data.

An intelligent learning approach for improving ECG signal classification and arrhythmia analysis.

The recognition of cardiac arrhythmia in minimal time is important to prevent sudden and untimely deaths. The proposed work includes a complete framework for analyzing the Electrocardiogram (ECG) signal. The three phases of analysis include 1) the ECG signal quality enhancement through noise suppression by a dedicated filter combination; 2) the feature extraction by a devoted wavelet design and 3) a proposed hidden Markov model (HMM) for cardiac arrhythmia classification into Normal (N), Right Bundle Branch Block (RBBB), Left Bundle Branch Block (LBBB), Premature Ventricular Contraction (PVC) and Atrial Premature Contraction (APC). The main features extracted in the proposed work are minimum, maximum, mean, standard deviation, and median. The experiments were conducted on forty-five ECG records in MIT BIH arrhythmia database and in MIT BIH noise stress test database. The proposed model has an overall accuracy of 99.7 % with a sensitivity of 99.7 % and a positive predictive value of 100 %. The detection error rate for the proposed model is 0.0004. This paper also includes a study of the cardiac arrhythmia recognition using an IoMT (Internet of Medical Things) approach.

Optimal Multi-Stage Arrhythmia Classification Approach.

Arrhythmia constitutes a problem with the rate or rhythm of the heartbeat, and an early diagnosis is essential for the timely inception of successful treatment. We have jointly optimized the entire multi-stage arrhythmia classification scheme based on 12-lead surface ECGs that attains the accuracy performance level of professional cardiologists. The new approach is comprised of a three-step noise reduction stage, a novel feature extraction method and an optimal classification model with finely tuned hyperparameters. We carried out an exhaustive study comparing thousands of competing classification algorithms that were trained on our proprietary, large and expertly labeled dataset consisting of 12-lead ECGs from 40,258 patients with four arrhythmia classes: atrial fibrillation, general supraventricular tachycardia, sinus bradycardia and sinus rhythm including sinus irregularity rhythm. Our results show that the optimal approach consisted of Low Band Pass filter, Robust LOESS, Non Local Means smoothing, a proprietary feature extraction method based on percentiles of the empirical distribution of ratios of interval lengths and magnitudes of peaks and valleys, and Extreme Gradient Boosting Tree classifier, achieved an F1-Score of 0.988 on patients without additional cardiac conditions. The same noise reduction and feature extraction methods combined with Gradient Boosting Tree classifier achieved an F1-Score of 0.97 on patients with additional cardiac conditions. Our method achieved the highest classification accuracy (average 10-fold cross-validation F1-Score of 0.992) using an external validation data, MIT-BIH arrhythmia database. The proposed optimal multi-stage arrhythmia classification approach can dramatically benefit automatic ECG data analysis by providing cardiologist level accuracy and robust compatibility with various ECG data sources.

Transfer Learning in ECG Classification from Human to Horse Using a Novel Parallel Neural Network Architecture.

Automatic or semi-automatic analysis of the equine electrocardiogram (eECG) is currently not possible because human or small animal ECG analysis software is unreliable due to a different ECG morphology in horses resulting from a different cardiac innervation. Both filtering, beat detection to classification for eECGs are currently poorly or not described in the literature. There are also no public databases available for eECGs as is the case for human ECGs. In this paper we propose the use of wavelet transforms for both filtering and QRS detection in eECGs. In addition, we propose a novel robust deep neural network using a parallel convolutional neural network architecture for ECG beat classification. The network was trained and tested using both the MIT-BIH arrhythmia and an own made eECG dataset with 26.440 beats on 4 classes: normal, premature ventricular contraction, premature atrial contraction and noise. The network was optimized using a genetic algorithm and an accuracy of 97.7% and 92.6% was achieved for the MIT-BIH and eECG database respectively. Afterwards, transfer learning from the MIT-BIH dataset to the eECG database was applied after which the average accuracy, recall, positive predictive value and F1 score of the network increased with an accuracy of 97.1%.

An Improved Convolutional Neural Network Based Approach for Automated Heartbeat Classification.

With age, our blood vessels are prone to aging, which induces cardiovascular disease. As an important basis for diagnosing heart disease and evaluating heart function, the electrocardiogram (ECG) records cardiac physiological electrical activity. Abnormalities in cardiac physiological activity are directly reflected in the ECG. Thus, ECG research is conducive to heart disease diagnosis. Considering the complexity of arrhythmia detection, we present an improved convolutional neural network (CNN) model for accurate classification. Compared with the traditional machine learning methods, CNN requires no additional feature extraction steps due to the automatic feature processing layers. In this paper, an improved CNN is proposed to automatically classify the heartbeat of arrhythmia. Firstly, all the heartbeats are divided from the original signals. After segmentation, the ECG heartbeats can be inputted into the first convolutional layers. In the proposed structure, kernels with different sizes are used in each convolution layer, which takes full advantage of the features in different scales. Then a max-pooling layer followed. The outputs of the last pooling layer are merged and as the input to fully-connected layers. Our experiment is in accordance with the AAMI inter-patient standard, which included normal beats (N), supraventricular ectopic beats (S), ventricular ectopic beats (V), fusion beats (F), and unknown beats (Q). For verification, the MIT arrhythmia database is introduced to confirm the accuracy of the proposed method, then, comparative experiments are conducted. The experiment demonstrates that our proposed method has high performance for arrhythmia detection, the accuracy is 99.06%. When properly trained, the proposed improved CNN model can be employed as a tool to automatically detect different kinds of arrhythmia from ECG.

An Effective LSTM Recurrent Network to Detect Arrhythmia on Imbalanced ECG Dataset.

To reduce the high mortality rate from cardiovascular disease (CVD), the electrocardiogram (ECG) beat plays a significant role in computer-aided arrhythmia diagnosis systems. However, the complex variations and imbalance of ECG beats make this a challenging issue. Since ECG beat data exist in heavily imbalanced category, an effective long short-term memory (LSTM) recurrence network model with focal loss (FL) is proposed. For this purpose, the LSTM network can disentangle the timing features in complex ECG signals, while the FL is used to resolve the category imbalance by downweighting easily identified normal ECG examples. The advantages of the proposed network have been verified in the MIT-BIH arrhythmia database. Experimental results show that the LSTM network with FL achieved a reliable solution to the problem of imbalanced datasets in ECG beat classification and was not sensitive to quality of ECG signals. The proposed method can be deployed in telemedicine scenarios to assist cardiologists into more accurately and objectively diagnosing ECG signals.

[Heartbeat-based end-to-end classification of arrhythmias].

OBJECTIVE: We propose a heartbeat-based end-to-end classification of arrhythmias to improve the classification performance for supraventricular ectopic beat (SVEB) and ventricular ectopic beat (VEB). METHODS: The ECG signals were preprocessed by heartbeat segmentation and heartbeat alignment. An arrhythmia classifier was constructed based on convolutional neural network, and the proposed loss function was used to train the classifier. RESULTS: The proposed algorithm was verified on MIT-BIH arrhythmia database. The AUC of the proposed loss function for SVEB and VEB reached 0.77 and 0.98, respectively. With the first 5 min segment as the local data, the diagnostic sensitivities for SVEB and VEB were 78.28% and 98.88%, respectively; when 0, 50, 100, and 150 samples were used as the local data, the diagnostic sensitivities for SVEB and VEB reached 82.25% and 93.23%, respectively. CONCLUSIONS: The proposed method effectively reduces the negative impact of class-imbalance and improves the diagnostic sensitivities for SVEB and VEB, and thus provides a new solution for automatic arrhythmia classification.

Arrhythmias in hematopoietic stem cell transplantation: A systematic review and meta-analysis.

BACKGROUND: There are controversial data regarding the relationship between hematopoietic stem cell transplantation and arrhythmias. This meta-analysis was performed to evaluate the incidence of arrhythmias in patients following hematopoietic stem cell transplantation (HSCT). METHODS: A literature search was conducted utilizing MEDLINE, EMBASE, and Cochrane Databases from inception through April 2019. Pooled incidence with 95% confidence interval (CI) were calculated using random-effects meta-analysis. The protocol for this meta-analysis is registered with PROSPERO (International Prospective Register of Systematic Reviews; no. CRD42019131833). RESULTS: Thirteen studies consisting of 10,587 patients undergoing HSCT were enrolled in this systematic review. Overall, the pooled estimated incidence of all types of arrhythmias following HSCT was 7.2% (95% CI: 4.9%-10.5%). With respect to the most common type of arrhythmia, the pooled estimated incidence of atrial fibrillation/atrial flutter (AF/AFL) within 30 days following HSCT was 4.2% (95% CI: 1.7%-9.6%). Egger's regression test demonstrated no significant publication bias in this meta-analysis of post-HSCT arrhythmia incidence. CONCLUSION: The overall estimated incidence of arrhythmias following HSCT was 7.2%. Future large scale studies are needed to further elucidate the significance and clinical impact of arrhythmias in post-HSCT patients.

[Automatic classification method of arrhythmia based on discriminative deep belief networks].

Existing arrhythmia classification methods usually use manual selection of electrocardiogram (ECG) signal features, so that the feature selection is subjective, and the feature extraction is complex, leaving the classification accuracy usually affected. Based on this situation, a new method of arrhythmia automatic classification based on discriminative deep belief networks (DDBNs) is proposed. The morphological features of heart beat signals are automatically extracted from the constructed generative restricted Boltzmann machine (GRBM), then the discriminative restricted Boltzmann machine (DRBM) with feature learning and classification ability is introduced, and arrhythmia classification is performed according to the extracted morphological features and RR interval features. In order to further improve the classification performance of DDBNs, DDBNs are converted to deep neural network (DNN) using the Softmax regression layer for supervised classification in this paper, and the network is fine-tuned by backpropagation. Finally, the Massachusetts Institute of Technology and Beth Israel Hospital Arrhythmia Database (MIT-BIH AR) is used for experimental verification. For training sets and test sets with consistent data sources, the overall classification accuracy of the method is up to 99.84% ± 0.04%. For training sets and test sets with inconsistent data sources, a small number of training sets are extended by the active learning (AL) method, and the overall classification accuracy of the method is up to 99.31% ± 0.23%. The experimental results show the effectiveness of the method in arrhythmia automatic feature extraction and classification. It provides a new solution for the automatic extraction of ECG signal features and classification for deep learning.