Abnormal Electrocardiogram Signals Research Articles

Advancing cardiac diagnostics: Exceptional accuracy in abnormal ECG signal classification with cascading deep learning and explainability analysis

Arrhythmias, cardiac rhythm disorders, demand precise diagnosis for effective treatment planning, emphasizing the crucial role of electrocardiogram (ECG) signal interpretation. While deep learning excels in ECG classification, its interpretability remains a challenge. Addressing this issue, our research introduces an innovative approach employing image-based ECG representations and cascading deep neural networks (CDNNs) to enhance arrhythmia detection precision. Initiating with the transformation of 12-lead ECG time-series data, particularly focusing on Lead II, into image-based representations using relative positioning matrices (RPM), this novel conversion captures spatial and temporal information concurrently, enriching the depth of our data analysis. Subsequently, CDNNs play a pivotal role in feature extraction, designed to automatically extract impactful features from image-based ECG representations, providing profound insights into cardiac electrical activity. Extracted features serve as inputs for the subsequent deep neural network for classification. Experimental results on the Shaoxing–Chapman ECG database, encompassing over 10,000 recordings with an 80/10/10 split, 80/20 split, 60/40 split, and 10-fold cross-validation, showcase a remarkable 100% accuracy in distinguishing between seven and four different arrhythmia classes. Incorporating SHapley Additive exPlanations (SHAP) values, our methodology offers comprehensive insights into the model’s decision-making, ensuring interpretability. Additionally, Gradient-weighted Class Activation Mapping (Grad-CAM) aids feature visualization, elucidating key features driving predictions. By amalgamating image-based representation, CDNNs, and interpretability techniques, our research not only attains exceptional classification performance but also boosts model transparency and trustworthiness. These findings hold promise for advancing arrhythmia diagnosis and enhancing interpretability in deep learning-based medical applications.

Detection of cardiac abnormalities from 12-lead ecg using complex wavelet sub-band features

Aim of the study. This research endeavours to optimize cardiac anomaly detection by introducing a method focused on selecting the most effective Daubechis wavelet families. The principal aim is to differentiate between cardiac states that are normal and abnormal by utilizing longer electrocardiogram (ECG) signal events based on the Apnea ECG dataset. Apnea ECG is often used to detect sleep apnea, a sleep disorder characterized by repeated interruptions in breathing during sleep. By using machine learning methods, such as Principal Component Analysis (PCA) and different classifiers, the goal is to improve the precision of cardiac irregularity identification. Used method. To extract important statistical and sub-band information from lengthy ECG signal episodes, the study uses a novel method that combines discrete wavelet transform with Principal Component Analysis (PCA) for dimension reduction. The methodology focuses on successfully categorizing ECG signals by utilizing several classifiers, including multilayer perceptron (MLP) neural network, Ensemble Subspace K-Nearest Neighbour(KNN), and Ensemble Bagged Trees, together with varied Daubechis wavelet families (db2, db3, db4, db5, db6). Brief Description of Results. The results emphasize the importance of the chosen Daubechis wavelet family, db5, and its superiority in ECG representation. The method distinguishes normal and abnormal ECG signals well on the Physionet Apnea ECG database. The Neural Network-based method accurately recognizes 100% of healthy signals and 97.8% of problematic ones with 98.6% accuracy. Findings. The Ensemble Subspace K-Nearest Neighbour (KNN) and Ensemble Bagged Trees methods got 87.1% accuracy and 0.89 and 0.87 AOC curve values on this dataset, showing that the method works. Precision values of 0.96, 0.86, and 0.86 for MLP Neural Network, KNN Subspace, and Ensemble Bagged Trees confirm their robustness. These findings suggest wavelet families and machine learning can improve cardiac abnormality detection and categorization.

IoT-based health information system using MitApp for abnormal electrocardiogram signal monitoring

Information systems are currently developing very rapidly, and this is inseparable from the role of internet of things (IoT) technology, especially in the world of telemedicine. MitApp is an open-source application that can be used to monitor electrocardiogram (ECG) signals in real-time. The aim of this study is to develop an IoT-based ECG signal monitoring system that utilizes the MitApp application to detect abnormal ECG signals that are characterized by symptoms of cardiac arrhythmias. To process ECG signal data obtained from lead electrode results, the research method utilizes Arduino Uno as a microcontroller. The result is then displayed on the thin film transistor (TFT) layer using the Nextion module. The ESP32 module is used as a Wi-Fi module to send data to the MitApp app on a smartphone. The results showed that the results of the comparison test of ECG signal module data with ECG simulator tools with beats per minute values of 60, 80, 100, 120, and 140 obtained an error rate of 0.05. Based on these results, there is potential to develop this feature and integrate the system with the patient management system to improve the effectiveness of remote monitoring.

ECG SIGNAL QUANTITATIVE ANALYSIS BASED ON EXTREMUM ENERGY DECOMPOSITION METHOD

Quantitative analysis of electrocardiogram (ECG) signals plays a pivotal role in objectively and quantitatively assessing cardiac electrical activity. This paper presents an innovative approach for quantitative ECG signal analysis utilizing extremum energy decomposition (EED). The methodology encompasses multiple steps: acquisition of unknown ECG signal under specific time and sampling conditions, denoising of acquired ECG signals, and subsequent decomposition of denoised ECG signals into a set of extremum modal function components alongside a residual. The n extremum modal function components obtained effectively represent different frequency bands. By evaluating these n extremum modal function components, the presence and severity of abnormalities within the ECG signal can be determined. The results showcased the effectiveness of the method in accurately identifying abnormal ECG signals, and the technique demonstrated robustness against noise interference, enhancing its practical utility in clinical and diagnostic settings. This research contributes to the field of ECG analysis by offering a quantitative toolset that enhances the objectivity and accuracy of abnormality assessment in cardiac electrical activity.

Quantum conditional generative adversarial network based on patch method for abnormal electrocardiogram generation

To address the scarcity and class imbalance of abnormal electrocardiogram (ECG) databases, which are crucial in AI-driven diagnostic tools for potential cardiovascular disease detection, this study proposes a novel quantum conditional generative adversarial algorithm (QCGAN-ECG) for generating abnormal ECG signals. The QCGAN-ECG constructs a quantum generator based on patch method. In this method, each sub-generator generates distinct features of abnormal heartbeats in different segments. This patch-based generative algorithm conserves quantum resources and makes QCGAN-ECG practical for near-term quantum devices. Additionally, QCGAN-ECG introduces quantum registers as control conditions. It encodes information about the types and probability distributions of abnormal heartbeats into quantum registers, rendering the entire generative process controllable. Simulation experiments on Pennylane demonstrated that the QCGAN-ECG could generate completely abnormal heartbeats with an average accuracy of 88.8%. Moreover, the QCGAN-ECG can accurately fit the probability distribution of various abnormal ECG data. In the anti-noise experiments, the QCGAN-ECG showcased outstanding robustness across various levels of quantum noise interference. These results demonstrate the effectiveness and potential applicability of the QCGAN-ECG for generating abnormal ECG signals, which will further promote the development of AI-driven cardiac disease diagnosis systems. The source code is available at github.com/VanSWK/QCGAN_ECG.

AnoFed: Adaptive anomaly detection for digital health using transformer-based federated learning and support vector data description

In digital healthcare applications, anomaly detection is an important task to be taken into account. For instance, in ECG (Electrocardiogram) analysis, the aim is often to detect abnormal ECG signals that are considered outliers. For such tasks, it has been shown that deep learning models such as Autoencoders (AEs) and Variational Autoencoders (VAEs) can provide state-of-the-art performance. However, they suffer from certain limitations. For example, the trivial method of threshold selection does not perform well if we do not know the reconstruction loss distribution in advance. In addition, since healthcare applications rely on highly sensitive personal information, data privacy concerns can arise when data are collected and processed in a centralized machine-learning setting. Hence, in order to address these challenges, in this paper, we propose AnoFed, a novel framework for combining the transformer-based AE and VAE with the Support Vector Data Description (SVDD) in a federated setting. It can enhance privacy protection, improve the explainability of results and support adaptive anomaly detection. Using ECG anomaly detection as a typical application of the framework in healthcare, we conducted experiments to show that the proposed framework is not only effective (in terms of the detection performance) but also efficient (in terms of computational costs), compared with a number of state-of-the-art methods in the literature. AnoFed is very lightweight in terms of the number of parameters and computation, hence it can be used in applications with resource-constrained edge devices.

A Novel Algorithm for Automated Human Single-Lead ECG Pre-Annotation and Beat-to-Beat Separation for Heartbeat Classification Using Autoencoders

An electrocardiogram (ECG) is used to check the electrical activity of the heart over a limited short-term or long-term period. Short-term observations are often used in hospitals or clinics, whereas long-term observations (often called continuous or stream-like ECG observations) are used to monitor the heart’s electrical activity on a daily basis and during different daily activities, such as sleeping, running, eating, etc. ECG can reflect the normal sinus rhythm as well as different heart problems, which might vary from Premature Atrial Contractions (PAC) and Premature Ventricular Contractions (PVC), to Sinus Arrest and many other problems. In order to perform such monitoring on a daily basis, it is very important to implement automated solutions that perform most of the work of the daily ECG analysis and could alert the doctors in case of any problem, and could even detect the type of the problem in order for the doctors to have an immediate report about the patient’s health status. This paper aims to provide a workflow for abnormal ECG signals detection from different sources of digitized ECG signals, including ambulatory devices. We propose an algorithm for ECG pre-annotation and beat-to-beat separation for heartbeat classification using Autoencoders. The algorithm includes the training of different models for different types of abnormal ECG signals, and has shown promising results for normal sinus rhythm and PVC compared to other solutions. This solution is proposed for no-noise and noisy signals as well.

Abnormal ECG detection based on an adversarial autoencoder.

Automatic detection and alarm of abnormal electrocardiogram (ECG) events play an important role in an ECG monitor system; however, popular classification models based on supervised learning fail to detect abnormal ECG effectively. Thus, we propose an ECG anomaly detection framework (ECG-AAE) based on an adversarial autoencoder and temporal convolutional network (TCN) which consists of three modules (autoencoder, discriminator, and outlier detector). The ECG-AAE framework is trained only with normal ECG data. Normal ECG signals could be mapped into latent feature space and then reconstructed as the original ECG signal back in our model, while abnormal ECG signals could not. Here, the TCN is employed to extract features of normal ECG data. Then, our model is evaluated on an MIT-BIH arrhythmia dataset and CMUH dataset, with an accuracy, precision, recall, F1-score, and AUC of 0.9673, 0.9854, 0.9486, 0.9666, and 0.9672 and of 0.9358, 0.9816, 0.8882, 0.9325, and 0.9358, respectively. The result indicates that the ECG-AAE can detect abnormal ECG efficiently, with its performance better than other popular outlier detection methods.

Machine Learning Techniques Applied for Exploring Heart Disease and Classifying Stages through ECG Signal

Electrocardiogram (ECG) has given crucial data concerning various heart attack criterion of the human heart this examination can be the key goal of the research for the detection and prevention of cardiac circumstances frighten. The proposed method using machine learning techniques for classifying and analyzing ECG signal processing and this research mainly developed for early discovery of heart diseases and also the prediction of stages level. The dataset was utilized as a person ECG signal of Heart Database which was taken from the UCI repository of Machine learning dataset vault. In this paper, a simple algorithm is presented for to discover R-peaks automatically from a single lead digital ECG data. The proposed method detecting the time interval of the ECG signal from the R-peaks level next level with the double squared difference signal is used to localize the region of QRS which is the time interval between the binary data. this method consists of different stages of sorting from the raw data for reducing nosier signal, threshold a difference signal of ECG by analyzing the time interval of QRS, and finally a comparison of relative magnitude to detect the region of interval processing to analyze accuracy result. The proposed research novel machine learning techniques of the multi-module neural network system (MMNNS) is used to analyze the imbalance problem form the ECG signal classification if the wave was abnormal then the user of dataset patients will be affected by heart diseases. Using the time interval varies from the range of QRS then it analyzes the abnormal of ECG by MMNNS algorithm to define the classification result and finally analyze the stages. If the patients were presented an abnormal ECG signal compared with a normal ECG signal wave graph then they were affected by heart diseases if whether they patients were affected then finally classify the stages of heart diseases whether they are predictable/unpredictable stages. The examination result is completed among two strategies on the better accuracy premise and effective training time of the process.

A Novel and Low Processing Time ECG Security Method Suitable for Sensor Node Platforms

An anonymisation of electrocardiogram (ECG) signal is essential during the distribution and storage in a public repository. In this paper, we propose a novel low processing time ECG anonymisation method based on the fast Fourier transform (FFT) algorithm that is suitable for sensor node platforms. The proposed framework was developed to address two major constraints in the Internet of Medical Thing environment, i.e., immediate need for securing ECG signal and efficient method for overcoming physical limitation of sensor nodes. Performance evaluation by way of computer simulation over normal and abnormal ECG signals showed that the proposed framework was able to conceal fiducial and non-fiducial features of the ECG signals. Additionally, it showed that the proposed framework offered flexibility in determining the secret key length of the anonymised ECG signal. Strong cross-correlation indicated close similarity between the original and the reconstructed ECG signals implying lossless reconstruction of the original ECG signal. Furthermore, the proposed method achieved a lower processing time security algorithm as compared with the recently proposed wavelet based anonymisation methods.

Premature Ventricular Contraction Recognition Based on a Deep Learning Approach

Electrocardiogram signal (ECG) is considered a significant biological signal employed to diagnose heart diseases. An ECG signal allows the demonstration of the cyclical contraction and relaxation of human heart muscles. This signal is a primary and noninvasive tool employed to recognize the actual life threat related to the heart. Abnormal ECG heartbeat and arrhythmia are the possible symptoms of severe heart diseases that can lead to death. Premature ventricular contraction (PVC) is one of the most common arrhythmias which begins from the lower chamber of the heart and can cause cardiac arrest, palpitation, and other symptoms affecting all activities of a patient. Nowadays, computer-assisted techniques reduce doctors' burden to assess heart arrhythmia and heart disease automatically. In this study, we propose a PVC recognition based on a deep learning approach using the MIT-BIH arrhythmia database. Firstly, 10 heartbeat and statistical features including three morphological features (RS amplitude, QR amplitude, and QRS width) and seven statistical features are computed for each signal. The extraction process of these features is conducted for 20 s of ECG data that create a feature vector. Next, these features are fed into a convolutional neural network (CNN) to find unique patterns and classify them more effectively. The obtained results prove that our pipeline improves the diagnosis performance more effectively.

Cancelable electrocardiogram biometric system based on chaotic encryption using three‐dimensional logistic map for biometric‐based cloud services

AbstractBiometric features are extensively employed for security purposes, but they are vulnerable to threats and can be lost or compromised. Electrocardiogram (ECG) has been utilized as one of the most favorable biometrics. However, it contains confidential patient health information and personal identification details. Moreover, security flaws take place in hacking scenarios. Therefore, original biometrics must be secured by preventing them from being used in biometric databases. The enhanced security trend in biometric authentication is cancelable biometrics. Cancelable biometric systems can be built by generating regularly repeated distortions of biometric features to secure the sensitive data of the users. When cancelable features are disclosed, distortion parameters are modified, and new cancelable templates are generated. This paper presents a proposed cancelable ECG recognition system based on the 3D chaotic logistic map encryption, which has highly efficient random characteristics with confusion and diffusion properties. Normal and abnormal ECG signals have been used to test the proposed cancelable biometric recognition system, and good access results have been obtained. The results of the simulation indicate that the proposed system is secure, reliable, and practicable even with ECG signals for patients.

Bidirectional Recurrent Neural Network Based Early Prediction of Cardiovascular Diseases using Electrocardiogram Signals for Type 2 Diabetic Patients

Introduction: The electrocardiogram (ECG) signal is important for early diagnosis of heart abnormalities. Type 2 diabetic individuals’ ECG signals provide pertinent data about their heart and are one of the most important diagnostic techniques used by doctors to identify Cardiovascular Disease (CVD). Bidirectional Recurrent Neural Network (RNN) classifies the features linked to normal and abnormal stage ECG signal. Aim: To analyse ECG signals of type 2 diabetic patients for early prediction of CVDs using feature extraction and bidirectional RNN based classification. Materials and Methods: This was a secondary data-based modelling study at Shri Ramasamy Memorial University Sikkim, India from December 2020 to January 2022. Different noises were removed by hybrid preprocessing filter made up of a Median and Savitzky-Golay filter. Undecimated Dual Tree Complex Wavelet Transform (UDTCWT) along with Detrended fluctuation (DA) analysis and Empirical Orthogonal Function (EOF) analysis were then used to extract features. These features were classified with Bidirectional RNN. Results: The proposed method was tested on the MIT-BIH, Physionet and DICARDIA databases, and the findings showed that it achieves an average accuracy of 97.6% when compared to the conventional techniques. Conclusion: The proposed method proves to be the most effective way for detecting anomalies in ECG signals in both the early and pathological stages. This method is also effective to diagnose the early intervention of cardiovascular symptoms.

Intracavitary Electrocardiogram Guidance Aids Excavation of Rhythm Abnormalities in Patients with Occult Heart Disease.

In this paper, the analysis of intracavitary electrocardiograms is used to guide the mining of abnormal cardiac rhythms in patients with hidden heart disease, and the algorithm is improved to address the data imbalance problem existing in the abnormal electrocardiogram signals, and a weight-based automatic classification algorithm for deep convolutional neural network electrocardiogram signals is proposed. By preprocessing the electrocardiogram data from the MIT-BIH arrhythmia database, the experimental dataset training algorithm model is obtained, and the algorithm model is migrated into the project. In terms of system design and implementation, by comparing the advantages and disadvantages of the electrocardiogram monitoring system platform, the overall design of the system was carried out in terms of functional and performance requirements according to the system realization goal, and a mobile platform system capable of classifying common abnormal electrocardiogram signals was developed. The system is capable of long-term monitoring and can invoke the automatic classification algorithm model of electrocardiogram signals for analysis. In this paper, the functional logic test and performance test were conducted on the main functional modules of the system. The test results show that the system can run stably and monitor electrocardiogram signals for a long time and can correctly call the deep convolutional neural network-based automatic electrocardiogram signal classification algorithm to analyze the electrocardiogram signals and achieve the requirements of displaying the electrocardiogram signal waveform, analyzing the heartbeat type, and calculating the average heart rate, which achieves the goal of real-time continuous monitoring and analysis of the electrocardiogram signals.

Evaluating Morphological Features of Electrocardiogram Signals for Diagnosing of Myocardial Infarction Using Classification-Based Feature Selection.

Background:Cardiovascular disease (CVD) is the first cause of world death, and myocardial infarction (MI) is one of the five primary disorders of CVDs which the patient electrocardiogram (ECG) analysis plays a dominant role in MI diagnosis. This research aims to evaluate some extracted features of ECG data to diagnose MI.Methods:In this paper, we used the Physikalisch-Technische Bundesanstalt database and extracted some morphological features, such as total integral of ECG, integral of the T-wave section, integral of the QRS complex, and J-point elevation from a cycle of normal and abnormal ECG waveforms. Since the morphology of healthy and abnormal ECG signals is different, we applied integral to different ECG cycles and intervals. We executed 100 of iterations on a 10-fold and 5-fold cross-validation method and calculated the average of statistical parameters to show the performance and stability of four classifiers, namely logistic regression (LR), simple decision tree, weighted K-nearest neighbor, and linear support vector machine. Furthermore, different combinations of proposed features were employed as a feature selection procedure based on classifier's performance using the aforementioned trained classifiers.Results:The results of our proposed method to diagnose MI utilizing all the proposed features with an LR classifier include 90.37%, 94.87%, and 86.44% for accuracy, sensitivity, specificity, respectively. Also, we calculated the standard deviation value for the accuracy of 0.006.Conclusion:Our proposed classification-based method successfully classified and diagnosed MI using different combinations of presented features. Consequently, all proposed features are valuable in MI diagnosis and are praiseworthy for future works.

Automated ECG classification based on 1D deep learning network.

The standard 12-lead electrocardiogram (ECG) records the heart's electrical activity from electrodes on the skin, and is widely used in screening and diagnosis of the cardiac conditions due to its low price and non-invasive characteristics. Manual examination of ECGs requires professional medical skills, and is strenuous and time consuming. Recently, deep learning methodologies have been successfully applied in the analysis of medical images. In this paper, we present an automated system for the identification of normal and abnormal ECG signals. A multi-channel multi-scale deep neural network (DNN) model is proposed, which is an end-to-end structure to classify the ECG signals without any feature extraction. Convolutional layers are used to extract primary features, and long short-term memory (LSTM) and attention are incorporated to improve the performance of the DNN model. The system was developed with a 12-lead ECG dataset provided by the Kaohsiung Medical University Hospital (KMUH). Experimental results show that the proposed system can yield high recognition rates in classifying normal and abnormal ECG signals.

Denoising of Electrocardiogram Signal Using S-Transform Based Time–Frequency Filtering Approach

Electrocardiogram (ECG) signals are damaged by various types of noise during acquisition and transmission which may mislead the analysis. In this paper, an automated denoising technique based on time–frequency filtering approach is proposed. The S-transform based time–frequency method with morphological processing is employed to visualize the spectrum of the ECG signal. The time–frequency plane is surface fitted to estimate the noise and then a threshold is used to eliminate it. The proposed method has been assessed with numerous abnormal and normal ECG signals selected from the MIT-BIH normal sinus rhythm database. Several noises with varying signal-to-noise ratio are considered for the simulation study. The results showed that the proposed technique is superior to the existing wavelet-based approach. It significantly reduces the mean square error, percentage root mean square difference and improves the signal-to-noise ratio (SNR). Moreover, at lower SNR condition, the proposed approach efficiently suppresses the noise. In the proposed approach, the requirement of the reference signal is eliminated; and at the same time, the structural information is preserved in the denoised signal.

Embedded Algorithm for QRS Detection Based on Signal Shape

Electrocardiogram (ECG) is one of the most useful medical examinations for the monitoring of cardiovascular diseases. The position and the duration of the QRS complex on the ECG signal are very important in the diagnosis of these diseases. Even though several R-peak (and hence QRS complex) detection algorithms are available, most are based on complex computations that require off-line processing on a personal computer (PC). However, with the advances in wearable devices and telemedicine, an algorithm that can run efficiently on a microcontroller (or embedded system) is needed. This article presents the development of an embedded algorithm for the detection of the QRS complex of an ECG signal. The algorithm is based on the shape and appearance of the signal. It extracts certain characteristics like the shape of the QRS complex, its slope, trend, and the duration between two successive QRS complexes, and then use them to increase detection accuracy. First, the R-peak is detected through the application of three levels of tests using adaptive thresholds. Second, from each R position, the positions of Q and S are detected using three other tests. To evaluate the performance of the algorithm, the MIT-BIH database was used and the sensitivity, positive prediction, and F1 score were used as evaluation metrics. The algorithm obtained average F1 scores of 99.67%, 99.73%, and 99.83% for the MIT-BIH Arrhythmia, Pacemaker Rhythm, and the Normal Sinus Rhythm Databases, respectively. Both normal and abnormal ECG signals were used in this performance test. The algorithm was then implemented on a microcontroller system, and its accuracy and run time were evaluated. The obtained F1 score results were the same as on the personal computer (PC) and an average run time of <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$16.23~\mu \text{s}$ </tex-math></inline-formula> per sample was obtained. The performance of the algorithm was also compared to other commonly used algorithms. The proposed algorithm has great potential in wearable systems for long-term monitoring.

Electrocardiogram signal classification for automated delineation using bidirectional long short-term memory

Analysis of electrocardiogram (ECG) signals is challenging due to the complexity of their signal morphology. Any irregularity in a cardiac rhythm can change the ECG waveform. A reliable machine learning model is developed here to provide substantial input to cardiologists and help confirm their diagnoses. To achieve high diagnostic accuracy, nearly all ECG analytics tools require records of the positions and morphologies of various segments of P-waves, QRS complexes, and T-waves to be kept in ECG records. However, analyzing such vast amounts of ECG records is not always easy. In most cases, it is challenging and highly time-consuming. Hence, an in-depth investigation regarding automatic ECG signal delineation is necessary. This paper proposes an automated delineation algorithm for ECG waveform signals that utilizes recurrent neural networks (RNNs) with bidirectional long short-term memory (LSTM) architecture. This delineation process consists of four steps: noise cancellation, ECG waveform segmentation, ECG signal classification in four classes (P-wave, QRS complex, T-wave, and isoelectric line) and model evaluation. The classification is conducted based on time duration, and each waveform is determined by using annotated data from the well-known QT database (QTDB). The results show that the proposed model produces satisfactory performance for the four classes in terms of average accuracy, sensitivity, specificity, precision, and F1-score, with values of 99.64%, 98.74%, 99.75%, 98.81%, and 98.78%, respectively. The proposed model is validated with abnormal ECG signals from the QTDB, i.e., MIT-BIH Arrhythmia, MIT-BIH ST Change, MIT-BIH Supraventricular Arrhythmia, European ST-T, and MIT-BIH Long-Term ECG. The results show that bidirectional LSTM can delineate ECG signals from QTDB in both normal and abnormal conditions. The proposed delineation method could be utilized in potential applications following further investigation.

Automated Diagnosis System for Outpatients and Inpatients With Cardiovascular Diseases

The identification of heart related diseases is challenging due to several contributory factors associated with patients, medical staff or medical materials used for diagnosis. Electrocardiogram (ECG) signal represents the electrical activity of the heart. It is the most common method used to diagnose patients with cardiovascular abnormalities. The evaluation commonly practiced by trained physicians can be sometimes subjective, time consuming and related to the observer status. This subjectivity can be more critical due to the double signification of the recorded ECG signals, mainly frequency and duration. In our paper, we present a comparative study of different Artificial Intelligence (AI) approaches as a very relevant tool to assist and improve the accuracy of cardiovascular diseases diagnosis. These models are trained on an online available MIT-BIH arrhythmia, normal rhythm sinus and BIDMC congestive heart failure databases and tested on our own collected data consisting of more than 72000 samples recorded in accordance with patients suffering from the same pathologies. The abnormal ECG signals are judged abnormal by comparison with normal heart beats. The work consists of testing and evaluating the performance of trained support vector machine (SVM), convolutional neural networks (CNN), quadratic discriminant, k-nearest neighbors and naive Bayes as classifiers to correctly and efficiently classify newly unlabeled data. Further, methodology comprises continuous wavelet transform (CWT), discrete wavelet transforms (DWT), maximum overlap discrete transform (MODWT) and autoregressive modelling (AM) as feature extraction techniques. We tested the prelisted methods with principal component analysis (PCA) to evaluate the dimensionality reduction influence on the overall accuracy and runtime measures. The consistency of performance is evaluated using overall accuracy with confidence interval (CI), misclassification cost and runtime. The study resulted on an overall accuracy of 99.92% with a CI of 99.07–100% and 98.63% with a CI of 95.1%-100% using quadratic discriminant and KNN respectively, both with a certainty level of 99%. The developed approach is robust and accurate and can be used for automated diagnosis of cardiovascular diseases.