Electrocardiogram Signal Classification in the Diagnosis of Heart Disease Based on RBF Neural Network.

Heart disease is a common disease affecting human health. Electrocardiogram (ECG) classification is the most effective and direct method to detect heart disease, which is helpful to the diagnosis of most heart disease symptoms. At present, most ECG diagnosis depends on the personal judgment of medical staff, which leads to heavy burden and low efficiency of medical staff. Automatic ECG analysis technology will help the work of relevant medical staff. In this paper, we use the MIT-BIH ECG database to extract the QRS features of ECG signals by using the Pan-Tompkins algorithm. After extraction of the samples, K-means clustering is used to screen the samples, and then, RBF neural network is used to analyze the ECG information. The classifier trains the electrical signal features, and the classification accuracy of the final classification model can reach 98.9%. Our experiments show that this method can effectively detect the abnormality of ECG signal and implement it for the diagnosis of heart disease.

Novel feature extraction method for signal analysis based on independent component analysis and wavelet transform.

Feature extraction is an important part of data processing that provides a basis for more complicated tasks such as classification or clustering. Recently many approaches for signal feature extraction were created. However, plenty of proposed methods are based on convolutional neural networks. This class of models requires a high amount of computational power to train and deploy and large dataset. Our work introduces a novel feature extraction method that uses wavelet transform to provide additional information in the Independent Component Analysis mixing matrix. The goal of our work is to combine good performance with a low inference cost. We used the task of Electrocardiography (ECG) heartbeat classification to evaluate the usefulness of the proposed approach. Experiments were carried out with an MIT-BIH database with four target classes (Normal, Vestibular ectopic beats, Ventricular ectopic beats, and Fusion strikes). Several base wavelet functions with different classifiers were used in experiments. Best was selected with 5-fold cross-validation and Wilcoxon test with significance level 0.05. With the proposed method for feature extraction and multi-layer perceptron classifier, we obtained 95.81% BAC-score. Compared to other literature methods, our approach was better than most feature extraction methods except for convolutional neural networks. Further analysis indicates that our method performance is close to convolutional neural networks for classes with a limited number of learning examples. We also analyze the number of required operations at test time and argue that our method enables easy deployment in environments with limited computing power.

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.

Fragmented QRS as a marker of coronary atherosclerosis severity evaluated by the Gensini score.

BACKGROUND: QRS fragmentation (fQRS) represents a marker of local myocardial fibrosis, especially in patients with CAD (coronary artery disease). However, little is known about the association between fQRS and the severity of coronary atherosclerosis as defined by the Gensini score. OBJECTIVE: To identify the angiographic and echocardiographic characteristics of patients with suspected CAD depending on the location and extent of fQRS. METHODS: A total of 178 patients who underwent coronary angiography were included in the study. fQRS was defined as the presence of RSR' and/or notching of the R/S wave (if QRS<120ms) or≥2 notches of the R/S wave (if QRS≥120ms). All patients were divided into three groups: non-fQRS; fQRS in 1-2 and≥3 leads. RESULTS: Statistically significant differences in the LVEF (left ventricular ejection fraction, P=0.009) and the degree of coronary atherosclerosis severity (P=0.008) were found among 3 groups. The median Gensini score was 7 in non-fQRS group (minimal CAD) and >20 in other groups (severe CAD). Both the anterior and lateral fQRS groups had a lower LVEF compared to no fQRS (P=0.039 and P=0.01, respectively). The median Gensini score was significantly higher in case of the lateral fQRS (P=0.037). fQRS in≥1 lead was associated with coronary occlusion (OR 2.1, 95% CI: 1.1-4.1, P=0.038). CONCLUSIONS: The presence of fQRS, particularly in lateral leads, can be a useful noninvasive marker of severe coronary atherosclerosis. Patients with≥1 fragmented lead have a lower LVEF, a higher Gensini score and a two-fold increased risk of occlusion.

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%.

Transfer learning for ECG classification.

Remote monitoring devices, which can be worn or implanted, have enabled a more effective healthcare for patients with periodic heart arrhythmia due to their ability to constantly monitor heart activity. However, these devices record considerable amounts of electrocardiogram (ECG) data that needs to be interpreted by physicians. Therefore, there is a growing need to develop reliable methods for automatic ECG interpretation to assist the physicians. Here, we use deep convolutional neural networks (CNN) to classify raw ECG recordings. However, training CNNs for ECG classification often requires a large number of annotated samples, which are expensive to acquire. In this work, we tackle this problem by using transfer learning. First, we pretrain CNNs on the largest public data set of continuous raw ECG signals. Next, we finetune the networks on a small data set for classification of Atrial Fibrillation, which is the most common heart arrhythmia. We show that pretraining improves the performance of CNNs on the target task by up to [Formula: see text], effectively reducing the number of annotations required to achieve the same performance as CNNs that are not pretrained. We investigate both supervised as well as unsupervised pretraining approaches, which we believe will increase in relevance, since they do not rely on the expensive ECG annotations. The code is available on GitHub at https://github.com/kweimann/ecg-transfer-learning .

Importance Ranking of Electrocardiogram Rhythms: A Primer for Curriculum Development.

INTRODUCTION: Electrocardiogram interpretation is an essential skill for emergency and critical care nurses and physicians. There remains a gap in standardized curricula and evaluation strategies used to achieve and assess competence in electrocardiogram interpretation. The purpose of this study was to develop an importance ranking of the 120 American Heart Association electrocardiogram diagnostic labels with interdisciplinary perspectives to inform curriculum development. METHODS: Data for this mixed methods study were collected through focus groups and individual semi-structured interviews. A card sort was used to assign relative importance scores to all 120 American Heart Association electrocardiogram diagnostic labels. Thematic analysis was used for qualitative data on participants' rationale for the rankings. RESULTS: The 18 participants included 6 emergency and critical care registered nurses, 5 cardiologists, and 7 emergency medicine physicians. The 5 diagnoses chosen as the most important by all disciplines were ventricular tachycardia, ventricular fibrillation, atrial fibrillation, complete heart block, and normal electrocardiogram. The "top 20" diagnoses by each discipline were also reported. Qualitative thematic content analysis revealed that participants from all 3 disciplines identified skill in electrocardiogram interpretation as clinically imperative and acknowledged the importance of recognizing normal, life threatening, and time-sensitive electrocardiogram rhythms. Additional qualitative themes, identified by individual disciplines, were reported. DISCUSSION: This mixed-methods approach provided valuable interdisciplinary perspectives concerning electrocardiogram curriculum case selection and prioritization. Study findings can provide a foundation for emergency and critical care educators to create local ECG educational programs. Further work is recommended to validate the list amongst a larger population of emergency and critical care frontline nurses and physicians.

Classification of 12-lead ECGs: the PhysioNet/Computing in Cardiology Challenge 2020.

OBJECTIVE: Vast 12-lead ECGs repositories provide opportunities to develop new machine learning approaches for creating accurate and automatic diagnostic systems for cardiac abnormalities. However, most 12-lead ECG classification studies are trained, tested, or developed in single, small, or relatively homogeneous datasets. In addition, most algorithms focus on identifying small numbers of cardiac arrhythmias that do not represent the complexity and difficulty of ECG interpretation. This work addresses these issues by providing a standard, multi-institutional database and a novel scoring metric through a public competition: the PhysioNet/Computing in Cardiology Challenge 2020. APPROACH: A total of 66 361 12-lead ECG recordings were sourced from six hospital systems from four countries across three continents; 43 101 recordings were posted publicly with a focus on 27 diagnoses. For the first time in a public competition, we required teams to publish open-source code for both training and testing their algorithms, ensuring full scientific reproducibility. MAIN RESULTS: A total of 217 teams submitted 1395 algorithms during the Challenge, representing a diversity of approaches for identifying cardiac abnormalities from both academia and industry. As with previous Challenges, high-performing algorithms exhibited significant drops ([Formula: see text]10%) in performance on the hidden test data. SIGNIFICANCE: Data from diverse institutions allowed us to assess algorithmic generalizability. A novel evaluation metric considered different misclassification errors for different cardiac abnormalities, capturing the outcomes and risks of different diagnoses. Requiring both trained models and code for training models improved the generalizability of submissions, setting a new bar in reproducibility for public data science competitions.

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.

Evaluation of electrocardiographic findings before and after hemodialysis session.

Dialysis patients have higher rates of sudden cardiac death. The study of the electrocardiogram could identify patients at risk of developing rhythm disorders. The aim of this study was to evaluate the electrocardiographic findings before and after the hemodialysis (HD) session and to examine associations of clinical and serum electrolytes with electrocardiogram findings. We conducted a multicentric transversal study, including chronic HD patients during January 2018. Standard 12-lead electrocardiogram was recorded, before and after the HD session. A medical history was documented. It included age, gender, initial nephropathy, and comorbidities. Serum potassium and total serum calcium were measured before a routine HD session. Serum potassium was measured after HD session. Corrected QT for heart rate was calculated using Bazett's formula. The study included 66 patients. Nineteen patients (28.8%) had hyperkalemia before the HD session and 44 (66.7%) patients had hypokalemia after the HD session. Seventeen patients had prolonged QTc interval (25.7%). On multiple regression analysis, only the prolonged QTc interval was significantly correlated with the serum potassium (P = 0.046).When comparing the mean values of electrocardiogram parameters before and after the HD session, we noted a significant change of heart rate (P = 0.001), R wave (P = 0.016), T wave (P = 0.001), and T/R (P = 0.001) wave. Delta K+ did not correlate with the change in T wave amplitude (r = 0.23, P = 0.59), R wave amplitude (r = -0.16, P = 0.2), T/R wave (r = 0.055, P = 0.65), or QRS duration (r = 0.023, P = 0.85). Delta QTc was correlated to ΔK+. We conclude that usual electrographic manifestations of hyperkalemia are less pronounced in HD patients. Our results confirmed the unstable status of cardiac electrophysiology during HD session.

Evaluation of Tp-e interval, Tp-e/QT ratio and Tp-e/QTc ratio in blunt chest trauma patients.

BACKGROUND: After blunt chest trauma, life-threatening arrhythmias may occur in the early post-injury period, as well as a few days after the injury. This study aimed to evaluate the risk of arrhythmias in blunt chest trauma patients using Tp-e interval, Tp-e/QT ratio and Tp-e/QTc ratio. METHODS: In this study, patients who applied to the emergency department due to blunt chest trauma were examined prospectively. The 12-lead ECG was performed to both blunt chest trauma and control group. ECG measurements of QT and Tp-e intervals were performed from both groups. RESULTS: A total of 81 participants; 41 blunt chest trauma patients and 40 healthy volunteers were included in this study. Tpe, Tpe/QT, Tpe/QTc values were statistically significant in the trauma group compared to the control group (p<0.001). Although Tpe/QTc, max QT and min QT were statistically significant (p<0.05) in patients with a rib fracture, no difference was detected concerning Tpe, Tpe/QT compared to no-rib fracture group (p>0.05). CONCLUSION: Tp-e interval, Tp-e/QT ratio and Tp-e/QTc ratio in ECG predict the arrhythmias that may occur in blunt cardiac trauma, especially in blunt chest trauma patients.

[QT Interval and Its Prolongation - What Does It Mean?] / QT-Zeit ­ Was fange ich eigentlich damit an?

The assessment of the QT interval has been an integral part of ECG interpretation since the first descriptions of long QT syndrome by Wolff in 1950 and by Jervell and Lange-Nielsen in 1957. The correct measurement of the QT interval as well as a correct interpretation of the causes and of the clinical consequences of a QT prolongation, however, may be difficult even for trained internists and cardiologists. In this review, we give an overview on physiological determinants of cardiac repolarization, its marker in the surface ECG - the QT interval -, methods to correctly assess QT interval duration, causes for pathologically prolonged QT intervals, and resulting clinical consequences. A correct measurement of the QT interval should be performed by using the "tangent method", excluding possible U waves. A heart rate correction formula should be employed to determine the heart rate corrected QT interval (QTc).Many factors, which may prolong the QT interval, should be checked whenever a QTc prolongation is observed. These include drugs, electrolyte imbalances, hormonal influence, and comorbidities. The correct management of a patient with (genetically determined) LQTS starts with a risk stratification based on genotype, ECG phenotype, clinical history, age, sex, concomitant diseases, drug therapies, and family history for syncope or sudden cardiac death. The therapeutic approaches for LQTS are multimodal. Prevention is the basis of the therapy and includes avoiding known risk factors / and potentially QT-prolonging drugs, and a pharmacological treatment with non-selective beta blockers. According to the risk profile and to the patient's lifestyle the implantation of an ICD or a pacemaker should be considered.

An Automatic R and T Peak Detection Method Based on the Combination of Hierarchical Clustering and Discrete Wavelet Transform.

The detection and delineation of QRS-complexes and T-waves in Electrocardiogram (ECG) is an important task because these features are associated with the cardiac abnormalities including ventricular arrhythmias that may lead to sudden cardiac death. In this paper, we propose a novel method for the R-peak and the T-peak detection using hierarchical clustering and Discrete Wavelet Transform (DWT) from the ECG signal. In the first step, a template of the single ECG beat is identified. Secondly, all R-peaks are detected by using hierarchical clustering. Then, each corresponding T-wave boundary is delineated based on the template morphology. Finally, the determination of T wave peaks is achieved based on the Modulus-Maxima Analysis (MMA) of the DWT coefficients. We evaluated the algorithm by using all records from the MIT-BIH arrhythmia database and QT database. The R-peak detector achieved a sensitivity of 99.89%, a positive predictivity of 99.97% and 99.83% accuracy over the validation MIT-BIH database. In addition, it shows a sensitivity of 100%, a positive predictivity of 99.83% in manually annotated QT database. It also shows 99.92% sensitivity and 99.96% positive predictivity over the automatic annotated QT database. In terms of the T-peak detection, our algorithm is verified with 99.91% sensitivity and 99.38% positive predictivity in manually annotated QT database.

Inter-Patient ECG Classification With Symbolic Representations and Multi-Perspective Convolutional Neural Networks.

This paper presents a novel deep learning framework for the inter-patient electrocardiogram (ECG) heartbeat classification. A symbolization approach especially designed for ECG is introduced, which can jointly represent the morphology and rhythm of the heartbeat and alleviate the influence of inter-patient variation through baseline correction. The symbolic representation of the heartbeat is used by a multi-perspective convolutional neural network (MPCNN) to learn features automatically and classify the heartbeat. We evaluate our method for the detection of the supraventricular ectopic beat (SVEB) and ventricular ectopic beat (VEB) on MIT-BIH arrhythmia dataset. Compared with the state-of-the-art methods based on manual features or deep learning models, our method shows superior performance: the overall accuracy of 96.4%, F1 scores for SVEB and VEB of 76.6% and 89.7%, respectively. The ablation study on our method validates the effectiveness of the proposed symbolization approach and joint representation architecture, which can help the deep learning model to learn more general features and improve the ability of generalization for unseen patients. Because our method achieves a competitive inter-patient heartbeat classification performance without complex handcrafted features or the intervention of the human expert, it can also be adjusted to handle various other tasks relative to ECG classification.

I-Vector-Based Patient Adaptation of Deep Neural Networks for Automatic Heartbeat Classification.

Automatic classification of electrocardiogram (ECG) signals is important for diagnosing heart arrhythmias. A big challenge in automatic ECG classification is the variation in the waveforms and characteristics of ECG signals among different patients. To address this issue, this paper proposes adapting a patient-independent deep neural network (DNN) using the information in the patient-dependent identity vectors (i-vectors). The adapted networks, namely i-vector adapted patient-specific DNNs (iAP-DNNs), are tuned toward the ECG characteristics of individual patients. For each patient, his/her ECG waveforms are compressed into an i-vector using a factor analysis model. Then, this i-vector is injected into the middle hidden layer of the patient-independent DNN. Stochastic gradient descent is then applied to fine-tune the whole network to form a patient-specific classifier. As a result, the adaptation makes use of not only the raw ECG waveforms from the specific patient but also the compact representation of his/her ECG characteristics through the i-vector. Analysis on the hidden-layer activations shows that by leveraging the information in the i-vectors, the iAP-DNNs are more capable of discriminating normal heartbeats against arrhythmic heartbeats than the networks that use the patient-specific ECG only for the adaptation. Experimental results based on the MIT-BIH database suggest that the iAP-DNNs perform better than existing patient-specific classifiers in terms of various performance measures. In particular, the sensitivity and specificity of the existing methods are all under the receiver operating characteristic curves of the iAP-DNNs.

Cardiac mortality predictability of T-wave alternans in young ST-elevated myocardial infarction patients with preserved cardiac function.

OBJECTIVE: Primary prevention of sudden cardiac death in ST-elevation myocardial infarction (STEMI) is a complicated issue due to the highly heterogeneous population. The effect of T-wave alternans (TWA) on cardiac mortality has been examined in various populations, most often in patients with a high risk of fatal arrhythmia, such as patients with a low left ventricular ejection fraction (LVEF). The aim of the present study was to investigate the prevalence of TWA and its relationship to cardiac mortality in young STEMI patients with preserved LVEF. METHODS: A total of 108 STEMI patients with preserved cardiac function who were under the age of 45 and underwent single-vessel primary percutaneous coronary intervention were enrolled in this prospective study. Preserved cardiac function was defined as an LVEF of ≥50% as detected with echocardiography 24 to 72 hours after the procedure. The TWA test was performed approximately 1 year after the STEMI occurrence. TWA positivity was defined with a maximal voltage of >64 µV and a heart rate of 125 beats per minute, as in previous studies. The patients were followed up for 5 years and overall cardiac mortality was measured. RESULTS: There was a positive TWA finding in 24 patients (22.2%). There was no significant difference in the use of medications, traditional risk factors, or LVEF in those with TWA positivity. During a follow-up period of 5 years, 7 patients (6.5%) reached the endpoint. Patients with TWA positivity had 10.7 times greater odds for 5-year cardiac mortality, independent of other risk factors. CONCLUSION: Clinicians should consider using the TWA test in young STEMI patients, as TWA positivity may be associated with increased cardiac mortality in this population.

Assessment of Electrocardiogram Rhythms by GoogLeNet Deep Neural Network Architecture.

The aim of this study is to design GoogLeNet deep neural network architecture by expanding the kernel size of the inception layer and combining the convolution layers to classify the electrocardiogram (ECG) beats into a normal sinus rhythm, premature ventricular contraction, atrial premature contraction, and right/left bundle branch block arrhythmia. Based on testing MIT-BIH arrhythmia benchmark databases, the scope of training/test ECG data was configured by covering at least three and seven R-peak features, and the proposed extended-GoogLeNet architecture can classify five distinct heartbeats; normal sinus rhythm (NSR), premature ventricular contraction (PVC), atrial premature contraction (APC), right bundle branch block (RBBB), and left bundle brunch block(LBBB), with an accuracy of 95.94%, an error rate of 4.06%, a maximum sensitivity of 96.9%, and a maximum positive predictive value of 95.7% for judging a normal or an abnormal beat with considering three ECG segments; an accuracy of 98.31%, a sensitivity of 88.75%, a specificity of 99.4%, and a positive predictive value of 94.4% for classifying APC from NSR, PVC, APC beats, whereas the error rate for misclassifying APC beat was relative low at 6.32%, compared with previous research efforts.

Electrocardiogram Classification Based on Faster Regions with Convolutional Neural Network.

The classification of electrocardiograms (ECG) plays an important role in the clinical diagnosis of heart disease. This paper proposes an effective system development and implementation for ECG classification based on faster regions with a convolutional neural network (Faster R-CNN) algorithm. The original one-dimensional ECG signals contain the preprocessed patient ECG signals and some ECG recordings from the MIT-BIH database in this experiment. Each ECG beat of one-dimensional ECG signals was transformed into a two-dimensional image for experimental training sets and test sets. As a result, we classified the ECG beats into five categories with an average accuracy of 99.21%. In addition, we did a comparative experiment using the one versus rest support vector machine (OVR SVM) algorithm, and the classification accuracy of the proposed Faster R-CNN was shown to be 2.59% higher.

The Value of Baseline and Arrhythmic ECG in the Interpretation of Arrhythmic Mechanisms.

Owing to the rapid development of new electrophysiologic techniques, our understanding of arrhythmias and their underlying mechanisms has reached unprecedented levels. In some cases, baseline ECG alterations can be identified before arrhythmia development; early recognition of these alterations is of utmost importance to start appropriate preventive therapies and stratify the risk according to patients' outcomes. Hereby, we report a systematic revision of main baseline ECG abnormalities and their implications on clinical outcomes.