Electrocardiogram Feature Extraction Research Articles

Multi-Scale and Multi-Channel Information Fusion for Exercise Electrocardiogram Feature Extraction and Classification

Multi-Scale and Multi-Channel Information Fusion for Exercise Electrocardiogram Feature Extraction and Classification

Machine learning based techniques for ECG noise removal and feature extraction

Machine learning (ML) is being applied to all aspects of life with the development of artificial intelligence (AI). This paper explores the application of machine learning technology in electrocardiogram (ECG) analysis to diagnose and classify a patient's current cardiac disease, predict possible future diseases, and provide a personalised treatment plan. However, several challenges have been highlighted. First, individual ECG signals display variability, causing concern about effective diagnosis based on changing ECG data. Second, different diseases can produce similar ECG results, requiring powerful classification algorithms to accurately classify diseases. Finally, using patient information to predict the probability of future heart attacks is critical to developing appropriate prevention and treatment strategies. Overcoming these challenges could revolutionise the field of cardiology. It could enable precise and proactive medical intervention. The study highlights the potential of machine learning to improve cardiovascular care and personalised medicine and emphasises the importance of addressing key challenges to maximise its impact in clinical practice.

The Role of Artificial Intelligence in Diagnosing Heart Disease in Humans: A Review

The electrical activity of the heart and the electrocardiogram (ECG) signal are fundamentally related. In the study that has been published, the ECG signal has been examined and used for a number of applications. The monitoring of heart rate and the analysis of heart rhythm patterns, the detection and diagnosis of cardiac diseases, the identification of emotional states, and the use of biometric identification methods are a few examples of applications in the field. Several various phases may be involved in the analysis of electrocardiogram (ECG) data, depending on the type of study being done. Preprocessing, feature extraction, feature selection, feature modification, and classification are frequently included in these stages. Every stage must be finished in order for the analysis to go smoothly. Additionally, accurate success measures and the creation of an acceptable ECG signal database are prerequisites for the analysis of electrocardiogram (ECG) signals. Identification and diagnosis of various cardiac illnesses depend heavily on the ECG segmentation and feature extraction procedure. Electrocardiogram (ECG) signals are frequently obtained for a variety of purposes, including the diagnosis of cardiovascular conditions, the identification of arrhythmias, the provision of physiological feedback, the detection of sleep apnea, routine patient monitoring, the prediction of sudden cardiac arrest, and the creation of systems for identifying vital signs, emotional states, and physical activities. The ECG has been widely used for the diagnosis and prognosis of a variety of heart diseases. Currently, a range of cardiac diseases can be accurately identified by computerized automated reports, which can then generate an automated report. This academic paper aims to provide an overview of the most important problems associated with using deep learning and machine learning to diagnose diseases based on electrocardiography, as well as a review of research on these techniques and methods and a discussion of the major data sets used by researchers.

Analysis of ECG-based arrhythmia detection system using machine learning

The 3D Discrete Wavelet Transform (DWT) and Support Vector Machine (SVM) are used in this study to analyze and characterize Electrocardiogram (ECG) signals. This technique consists of three stages: ECG signal preprocessing, feature extraction, and ECG signal order. The 3D wavelet transform is a signal preprocessing technique, de-noising, along with wavelet coefficient extraction.•SVM is used to categorize the ECG through each of the nine heartbeat types recognized by the various classifiers. For this work, around 6400 ECG beats were looked at over the China Physiological Signal Challenge (CPSC) 2018 arrhythmia dataset.•The best degree of exactness was acquired when level 4 rough constants with Symlet-8 (Sym8) channel were utilized for arrangement. Utilizing the ECG signals from CPSC 2018 data set, the SVM classifier has a normal precision of 99.02%, which is much better than complex support vector machine (CSVM) 98.5%, and weighted support vector machine (WSVM) 99%.•The suggested approach is far superior to others in terms of accuracy, and classification of several diseases of arrhythmia.

A dynamic learning-based ECG feature extraction method for myocardial infarction detection

Objective. Myocardial infarction (MI) is one of the leading causes of human mortality in all cardiovascular diseases globally. Currently, the 12-lead electrocardiogram (ECG) is widely used as a first-line diagnostic tool for MI. However, visual inspection of pathological ECG variations induced by MI remains a great challenge for cardiologists, since pathological changes are usually complex and slight. Approach. To have an accuracy of the MI detection, the prominent features extracted from in-depth mining of ECG signals need to be explored. In this study, a dynamic learning algorithm is applied to discover prominent features for identifying MI patients via mining the hidden inherent dynamics in ECG signals. Firstly, the distinctive dynamic features extracted from the multi-scale decomposition of dynamic modeling of the ECG signals effectively and comprehensibly represent the pathological ECG changes. Secondly, a few most important dynamic features are filtered through a hybrid feature selection algorithm based on filter and wrapper to form a representative reduced feature set. Finally, different classifiers based on the reduced feature set are trained and tested on the public PTB dataset and an independent clinical data set. Main results. Our proposed method achieves a significant improvement in detecting MI patients under the inter-patient paradigm, with an accuracy of 94.75%, sensitivity of 94.18%, and specificity of 96.33% on the PTB dataset. Furthermore, classifiers trained on PTB are verified on the test data set collected from 200 patients, yielding a maximum accuracy of 84.96%, sensitivity of 85.04%, and specificity of 84.80%. Significance. The experimental results demonstrate that our method performs distinctive dynamic feature extraction and may be used as an effective auxiliary tool to diagnose MI patients.

Detection of arrhythmia in 12-lead varied-length ECG using multi-branch signal fusion network

Objective. Automatic detection of arrhythmia based on electrocardiogram (ECG) plays a critical role in early prevention and diagnosis of cardiovascular diseases. With the increase in widely available digital ECG data and the development of deep learning, multi-class arrhythmia classification based on automatic feature extraction of ECG has become increasingly attractive. However, the majority of studies cannot accept varied-length ECG signals and have limited performance in detecting multi-class arrhythmias. Approach. In this study, we propose a multi-branch signal fusion network (MBSF-Net) for multi-label classification of arrhythmia in 12-lead varied-length ECG. Our model utilizes the complementary power between different structures, which include Inception with depthwise separable convolution (DWS-Inception), spatial pyramid pooling (SPP) Layer, and multi-scale fusion Resnet (MSF-Resnet). The proposed method can extract features from each lead of 12-lead ECG recordings separately and then effectively fuse the features of each lead by integrating multiple convolution kernels with different receptive fields, which can achieve the information of complementation between different angles of the ECG signal. In particular, our model can accept 12-lead ECG signals of arbitrary length. Main results. The experimental results show that our model achieved an overall classification F1 score of 83.8% in the 12-lead ECG data of CPSC-2018. In addition, the F1 score of the MBSF-Net performed best among the MBF-Nets which are removed the SPP layer from MBSF-Net. In comparison with the latest ECG classification algorithms, the proposed model can be applied in varied-length signals and has an excellent performance, which not only can fully retain the integrity of the original signals, but also eliminates the cropping/padding signal beforehand when dealing with varied-length signal database. Significance. MBSF-Net provides an end-to-end multi-label classification model with outperfom performance, which allows detection of disease in varied-length signals without any additional cropping/padding. Moreover, our research is beneficial to the development of computer-aided diagnosis.

A deep learning approach to cardiovascular disease classification using empirical mode decomposition for ECG feature extraction

Deep learning has achieved promising results on a broad spectrum of tasks using an end-to-end approach, and domain-specific knowledge can be used to supplement it by either constraining the solution-space, or to transform data such that relevant signals are more influential during the training process. This is especially important in the context of continuous recordings or wearable-based monitoring, where signal quality may be constrained by hardware, such as a limited number of available electrodes, or by external perturbations. In this study, a global feature extraction method for multi-periodic signals is proposed, which effectively classifies multi-cycle normal and abnormal heartbeats. The abnormal heartbeats are classed as the Left Bundle Branch Block beat (LBBB), Right Bundle Branch Block beat (RBBB), and Paced beat (P). The signal analysis process of the electrocardiogram (ECG) recordings consist of three main stages: (I) segmentation of the ECG data and partitioning of the data set; (II) generation of an overall feature map representing the “heartbeat condition” based on Empirical Mode Decomposition (EMD); and (III) a classification stage for determining the patient’s heartbeat condition. The extracted feature image is used to classify the heartbeat condition using a two-dimensional Convolutional Neural Network (CNN). This method is applied to the publicly available MIT-BIH arrhythmia database. Experimental results show that the reconstructed ECG features outperform use of the raw ECG signals. Under self-test conditions of 3 min of ECG signal, the Total Classification Accuracy (TCA) was approximately 99.01%, while the TCA without the proposed method was approximately 90.69%.

Deep learning assessment of left ventricular hypertrophy based on electrocardiogram.

BackgroundCurrent electrocardiogram (ECG) criteria of left ventricular hypertrophy (LVH) have low sensitivity. Deep learning (DL) techniques have been widely used to detect cardiac diseases due to its ability of automatic feature extraction of ECG. However, DL was rarely applied in LVH diagnosis. Our study aimed to construct a DL model for rapid and effective detection of LVH using 12-lead ECG.MethodsWe built a DL model based on convolutional neural network-long short-term memory (CNN-LSTM) to detect LVH using 12-lead ECG. The echocardiogram and ECG of 1,863 patients obtained within 1 week after hospital admission were analyzed. Patients were evenly allocated into 3 sets at 3:1:1 ratio: the training set (n = 1,120), the validation set (n = 371) and the test set 1 (n = 372). In addition, we recruited 453 hospitalized patients into the internal test set 2. Different DL model of each subgroup was developed according to gender and relative wall thickness (RWT).ResultsThe LVH was predicted by the CNN-LSTM model with an area under the curve (AUC) of 0.62 (sensitivity 68%, specificity 57%) in the test set 1, which outperformed Cornell voltage criteria (AUC: 0.57, sensitivity 48%, specificity 72%) and Sokolow-Lyon voltage (AUC: 0.51, sensitivity 14%, specificity 96%). In the internal test set 2, the CNN-LSTM model had a stable performance in predicting LVH with an AUC of 0.59 (sensitivity 65%, specificity 57%). In the subgroup analysis, the CNN-LSTM model predicted LVH by 12-lead ECG with an AUC of 0.66 (sensitivity 72%, specificity 60%) for male patients, which performed better than that for female patients (AUC: 0.59, sensitivity 50%, specificity 71%).ConclusionOur study established a CNN-LSTM model to diagnose LVH by 12-lead ECG with higher sensitivity than current ECG diagnostic criteria. This CNN-LSTM model may be a simple and effective screening tool of LVH.

Convolutional Autoencoding and Gaussian Mixture Clustering for Unsupervised Beat-to-Beat Heart Rate Estimation of Electrocardiograms from Wearable Sensors.

Heart rate is one of the most important diagnostic bases for cardiovascular disease. This paper introduces a deep autoencoding strategy into feature extraction of electrocardiogram (ECG) signals, and proposes a beat-to-beat heart rate estimation method based on convolution autoencoding and Gaussian mixture clustering. The high-level heartbeat features were first extracted in an unsupervised manner by training the convolutional autoencoder network, and then the adaptive Gaussian mixture clustering was applied to detect the heartbeat locations from the extracted features, and calculated the beat-to-beat heart rate. Compared with the existing heartbeat classification/detection methods, the proposed unsupervised feature learning and heartbeat clustering method does not rely on accurate labeling of each heartbeat location, which could save a lot of time and effort in human annotations. Experimental results demonstrate that the proposed method maintains better accuracy and generalization ability compared with the existing ECG heart rate estimation methods and could be a robust long-time heart rate monitoring solution for wearable ECG devices.

Research on high-efficiency electrocardiogram automatic classification based on autoregressive moving average model fitting

The automatic detection of arrhythmia is of great significance for the early prevention and diagnosis of cardiovascular diseases. Traditional arrhythmia diagnosis is limited by expert knowledge and complex algorithms, and lacks multi-dimensional feature representation capabilities, which is not suitable for wearable electrocardiogram (ECG) monitoring equipment. This study proposed a feature extraction method based on autoregressive moving average (ARMA) model fitting. Different types of heartbeats were used as model inputs, and the characteristic of fast and smooth signal was used to select the appropriate order for the arrhythmia signal to perform coefficient fitting, and complete the ECG feature extraction. The feature vectors were input to the support vector machine (SVM) classifier and K-nearest neighbor classifier (KNN) for automatic ECG classification. MIT-BIH arrhythmia database and MIT-BIH atrial fibrillation database were used to verify in the experiment. The experimental results showed that the feature engineering composed of the fitting coefficients of the ARMA model combined with the SVM classifier obtained a recall rate of 98.2% and a precision rate of 98.4%, and the F 1 index was 98.3%. The algorithm has high performance, meets the needs of clinical diagnosis, and has low algorithm complexity. It can use low-power embedded processors for real-time calculations, and it's suitable for real-time warning of wearable ECG monitoring equipment.

Towards noninvasive and fast detection of Glycated hemoglobin levels based on ECG using convolutional neural networks with multisegments fusion and Varied-weight

Glycated hemoglobin A1c (HbA1c) is regarded as a gold standard to evaluate long-term blood glucose control, and it is also a crucial metric in diabetes screening, diagnosis, and management. However, thus far, the HbA1c measurement methods are invasive and painful. Considering that HbA1c levels are associated with cardiovascular autonomic neuropathy, in this paper, a novel Electrocardiogram (ECG)-based approach was presented for noninvasive and fast detection of HbA1c levels using 60-second, single-lead ECG waveform. For this purpose, a total of 317,105 ECG datasets encompassing 370 patients with diabetes were obtained using wearable devices. Furthermore, the ECG preprocessing was based on autocorrelation analysis. The convolutional neural networks with multisegment fusion and varied-weight (CNN-MFVW) were proposed to achieve ECG feature extraction and HbA1c detection. The results showed that the average accuracy, precision, recall, and F1-score of the proposed algorithm were 0.9015, 0.9051, 0.8991 and 0.9013 respectively. Moreover, the area under the curve (AUC) was up to 0.9899, which was higher than other algorithms of conventional CNN and CNN-LSTM. Therefore, we conclude that the proposed approach for noninvasive and fast detection of HbA1c levels has potentials in practical applications.

Representation and Denoising of ECG Signal Using Hybrid Filtering Approach

Electrocardiogram (ECG) signal plays an imperative role in monitoring and examining the health condition of the heart. ECG signal represents the electrical activity of the heat. The most consequential noises that degrade important features in ECG signal are powerline interference noise, external electromagnetic field interference noise, baseline wandering and electroencephalogram noise. The features of ECG signal obtained in time domain is not sufficient for analyzing the ECG signal. As the signal is non-stationary, the time-frequency representation can be used for feature extraction. The Short Time Fourier Transform can be used but its time frequency precision is not optimal. In this current project, we will be able to implement the ideology proposed to overcome the problem among various time frequency transformation. The discrete wavelet transform (DWT) is used which gives effective results for non-stationary signals like ECG signal which may be often contaminated. The combination of Savitzky-Golay filtering and DWT can be used for ECG denoising and feature extraction which has the advantage of preserving the important feature by elimination the noise components. The method is applied for the database which is taken from MIT- BIH arrhythmia and the algorithm is implemented in MATLAB platform.

A particle swarm optimization improved BP neural network intelligent model for electrocardiogram classification

BackgroundAs proven to reflect the work state of heart and physiological situation objectively, electrocardiogram (ECG) is widely used in the assessment of human health, especially the diagnosis of heart disease. The accuracy and reliability of abnormal ECG (AECG) decision depend to a large extent on the feature extraction. However, it is often uneasy or even impossible to obtain accurate features, as the detection process of ECG is easily disturbed by the external environment. And AECG got many species and great variation. What’s more, the ECG result obtained after a long time past, which can not reach the purpose of early warning or real-time disease diagnosis. Therefore, developing an intelligent classification model with an accurate feature extraction method to identify AECG is of quite significance. This study aimed to explore an accurate feature extraction method of ECG and establish a suitable model for identifying AECG and the diagnosis of heart disease.MethodsIn this research, the wavelet combined with four operations and adaptive threshold methods were applied to filter the ECG and extract its feature waves first. Then, a BP neural network (BPNN) intelligent model and a particle swarm optimization (PSO) improved BPNN (PSO-BPNN) intelligent model based on MIT-BIH open database was established to identify ECG. To reduce the complexity of the model, the principal component analysis (PCA) was used to minimize the feature dimension.ResultsWavelet transforms combined four operations and adaptive threshold methods were capable of ECG filtering and feature extraction. PCA can significantly deduce the modeling feature dimension to minimize the complexity and save classification time. The PSO-BPNN intelligent model was suitable for identifying five types of ECG and showed better effects while comparing it with the BPNN model.ConclusionIn summary, it was further concluded that the PSO-BPNN intelligent model would be a suitable way to identify AECG and provide a tool for the diagnosis of heart disease.

Independent component analysis with learning algorithm for electrocardiogram feature extraction and classification

Electrocardiogram (ECG) analysis is a conventional way of finding heart abnormality. It is a clinical procedure in which the electrical activity of the heart is measured during every cardiac cycle and checked for healthiness of the heart. It is approximated in this industrialized world that millions of people expire every 12 months because of various coronary heart diseases and short of prompt detection of uncharacteristic heart rhythms. To detect these abnormalities promptly, the ECG measures should provide the cardiac signals without any mixtures or other disturbances. Though accurate classification of ECG is a challenging task as it varies with time and also with persons of different ages, it is the need of the hour. In this proposed research work, an improved independent component analysis (ICA) algorithm is used to extract pure ECG components from the ECG mixtures before the signals are applied to machine learning classifiers for accurate detection and classification of ECG signals. These machine learning models are applied after the signals are preprocessed to reduce the dimensionality and the training time. This work also uses deep learning convolution neural network (CNN) model with different optimizers for ECG classification and analysis. Classification performance of these algorithms is improved when classification is done after extracting the features using ICA technique.

A Comprehensive Survey on ECG Signals as New Biometric Modality for Human Authentication: Recent Advances and Future Challenges

Electrocardiogram (ECG) has extremely discriminative characteristics in the biometric field and has recently received significant interest as a promising biometric trait. However, ECG signals are susceptible to several types of noises, such as baseline wander, powerline interference, and high/low-frequency noises, making it challenging to realize biometric identification systems precisely and robustly. Therefore, ECG signal denoising is a major preprocessing step and plays a crucial role in ECG-based biometric human identification. ECG signal analysis for biometric recognition can combine several steps, such as preprocessing, feature extraction, feature selection, feature transformation, and classification which is a very challenging task. Moreover, the employed success measures and appropriate constitution of the ECG signal database also play significant roles in biometric system analysis, considering that publicly available databases are essential by the research community to evaluate the performance of their proposed algorithms. In this survey, we review most of the techniques employed for the ECG as biometrics for human authentication. Firstly, we present an overview and discussion on ECG signal preprocessing, feature extraction, feature selection, and feature transformation for ECG-based biometric systems. Secondly, we present a survey of the available ECG databases to evaluate and compare the acquisition protocol, acquisition hardware, and acquisition resolution (bits) for ECG-based biometric systems. Thirdly, we also present a survey on different techniques, including deep learning methods: deep supervised learning, deep semi-supervised learning, and deep unsupervised learning, for ECG signal classification. Lastly, we present the state-of-art approaches of information fusion in multimodal biometric systems.

An Efficient Heartbeats Classifier Based on Optimizing Convolutional Neural Network Model

Recently, deep learning models have emerged as promising methods for the diagnosis of different diseases. Cardiac disease is among the leading life-threatening diseases on a global scale. The aim of this paper is to propose an optimized Convolutional Neural Network (CNN) model for the classification of electrocardiogram (ECG) heartbeat data. The proposed ECG classification approach is designed with an optimal CNN configuration to classify cardiac arrhythmias quickly and effectively. Finding an optimal configuration for the CNN hyperparameters is time-consuming and needs extensive experimentation. To overcome this challenge, we present an optimization step for the proposed CNN model using a customized genetic algorithm. It provides an automatic suggestion for the best hyperparameter settings of the proposed CNN. The challenge in utilizing the genetic algorithm is that its operators need to be customized to handle our problem domain. Our approach accepts raw ECG signals without any preprocessing steps, which has benefit in saving the computation time. Our approach also provides a resampling step to ensure generalization, to better handle imbalanced ECG classes. Experiments show promising results of our proposed approach against other approaches whose CNN hyperparameters setting depended on numerous trials, requiring extensive ECG feature extraction steps, and do not consider imbalanced classes. The performance of our proposed approach is better than other existing methods both in terms of higher classification accuracy (98.45%), and lower computational complexity.

Performance Analysis of ECG Arrhythmia Classification based on Different SVM Methods

Heart arrhythmias are the different types of heartbeats which are irregular in nature. In Tachycardia the heartbeat works too fast and in case of Bradycardia it works too slow. In the study of different cardiac conditions automatic detection of heart arrhythmia is done by the classification and feature extraction of Electrocardiogram(ECG) data. Various Support Vector Machine based methods are used to analyze and classify ECG signals for arrhythmia detection. There are several Support Vector Machine (SVM) methods used to classify the ECG data such as one against all, one against one and fuzzy decision function. This classification detects the existence of the arrhythmia and it helps the physicians to treat the heart patient with more accurate way. To train SVM, the MIT BIH Arrhythmia database is used which works with the heart disorder like sinus bradycardy, old inferior myocardial infarction, coronary artery disease, right bundle branch block. All three methods are implemented in proper way, and their rate of accuracy with SVM classifier is optimal when it is processed with the one-against-all method. The data sets of ECG arrhythmia are usually complex in nature, so for the SVM based classification one-against-all method has great impact and will fetch better result.

Healthcare knowledge of relationship between time series electrocardiogram and cigarette smoking using clinical records

BackgroundIn the few studies of clinical experience available, cigarette smoking may be associated with ischemic heart disease and acute coronary events, which can be reflected in the electrocardiogram (ECG). However, there is no formal proof of a significant relationship between cigarette smoking and electrocardiogram results. In this study, we therefore investigate and prove the relationship between electrocardiogram and smoking using unsupervised neural network techniques.MethodsIn this research, a combination of two techniques of pattern recognition; feature extraction and clustering neural networks, is specifically investigated during the diagnostic classification of cigarette smoking based on different electrocardiogram feature extraction methods, such as the reduced binary pattern (RBP) and Wavelet features. In this diagnostic system, several neural network models have been obtained from the different training subsets by clustering analysis. Unsupervised neural network of clustering cigarette smoking was then implemented based on the self-organizing map (SOM) with the best performance.ResultsTwo ECG datasets were investigated and analysed in this prospective study. One is the public PTB diagnostic ECG databset with 290 samples (age 17–87, mean 57.2; 209 men and 81 women; 73 smoking and 133 non-smoking). The other ECG database is from Taichung Veterans General Hospital (TVGH) and includes 480 samples (240 smoking, and 240 non-smoking). The diagnostic accuracy regarding smoking and non-smoking in the PTB dataset reaches 80.58% based on the RBP feature, and 75.63% in the second dataset based on Wavelet feature.ConclusionsThe electrocardiogram diagnostic system performs satisfactorily in the cigarette smoking habit analysis task, and demonstrates that cigarette smoking is significantly associated with the electrocardiogram.

Personal identity verification based ECG biometric using non-fiducial features

Biometrics was used as an automated and fast acceptable technology for human identification and it may be behavioral or physiological traits. Any biometric system based on identification or verification modes for human identity. The electrocardiogram (ECG) is considered as one of the physiological biometrics which impossible to mimic or stole. ECG feature extraction methods were performed using fiducial or non-fiducial approaches. This research presents an authentication ECG biometric system using non-fiducial features obtained by Discrete Wavelet Decomposition and the Euclidean Distance technique was used to implement the identity verification. From the obtained results, the proposed system accuracy is 96.66% also, using the verification system is preferred for a large number of individuals as it takes less time to get the decision.

Fast and Accurate Algorithm for ECG Authentication Using Residual Depthwise Separable Convolutional Neural Networks

The electrocardiogram (ECG) is relatively easy to acquire and has been used for reliable biometric authentication. Despite growing interest in ECG authentication, there are still two main problems that need to be tackled, i.e., the accuracy and processing speed. Therefore, this paper proposed a fast and accurate ECG authentication utilizing only two stages, i.e., ECG beat detection and classification. By minimizing time-consuming ECG signal pre-processing and feature extraction, our proposed two-stage algorithm can authenticate the ECG signal around 660 μs. Hamilton’s method was used for ECG beat detection, while the Residual Depthwise Separable Convolutional Neural Network (RDSCNN) algorithm was used for classification. It was found that between six and eight ECG beats were required for authentication of different databases. Results showed that our proposed algorithm achieved 100% accuracy when evaluated with 48 patients in the MIT-BIH database and 90 people in the ECG ID database. These results showed that our proposed algorithm outperformed other state-of-the-art methods.