ECG Signal Classification Research Articles

An Effective Classification of Heart Rate Data using PSO-FCM Clustering and Enhanced Support Vector Machine

Background/Objectives: Heart Rate Variability is an essential feature which decides the condition of human heart. ECG is used as diagnostic tool to access the electrical function of the heart. Methods/Statistical Analysis: The nine linear and nonlinear features are derived from the HRV signals. The feature extraction is carried out with the help of Particle Swarm Optimization (PSO) for data reduction. In proposed scheme Fuzzy C-Means (FCM) clustering and classifier integrated to enhance the accuracy result for ECG beat classification. Findings: The Enhanced SVM classifier classifies the heart rate data. Enhanced SVM classifier groups the linear and non-linear parameters as inputs, which are derived from the HRV signal. The denoise signals are classified and identifies the pattern for better classification of ECG signal. Application/Improvements: The proposed scheme is experimented with the assistance of the most commonly used MIT-BIH arrhythmia database and adequate results were obtained with an accuracy level of 98.38% than the other well-known approaches.

A new personalized ECG signal classification algorithm using Block-based Neural Network and Particle Swarm Optimization

Abstract The purpose of this paper is the classification of ECG heartbeats of a patient in five heartbeat types according to AAMI recommendation, using an artificial neural network. In this paper a Block-based Neural Network (BBNN) has been used as the classifier. The BBNN is created from 2-D array of blocks which are connected to each other. The internal structure of each block depends on the number of incoming and outgoing signals. The overall construction of the network is determined by the moving of signals through the network blocks. The Network structure and the weights are optimized using Particle Swarm Optimization (PSO) algorithm. The input of the BBNN is a vector which its elements are the features extracted from the ECG signals. In this paper Hermit function coefficient and temporal features which have been extracted from ECG signals, create the input vector of the BBNN. The BBNN parameters have been optimized by PSO algorithm which can overcome the possible changes of ECG signals from time-to-time and/or person-to-person variations. Therefore the trained BBNN has an unique structure for each person. The performance evaluation using the MIT-BIH arrhythmia database shows a high classification accuracy of 97%.

ENHANCEMENT OF ECG CLASSIFICATION USING GA AND PSO

ECG signal classification utilizes for different predictions of heart diseases. These ECG signals have to be classified using different frequency bands according to different energy levels for better prediction of features. These signals have to be classified in different five bands P, Q, R, S and T. These sub-bands provide peak information available in different sub-bands. For the classification various approaches have to be implemented for filtration of signal. In the purposed work Adaptive filter has been implemented for the noise reduction from these signals. Classification of the ECG signal has been optimized using Genetic Algorithm and Particle Swarm Optimization approach. These approaches of classification provide better results i.e. 100 and 100 for 106 and 119 respectively for energy levels of ECG signal.

Neuro-Fuzzy Based ECG Signal Classification with A Gaussian Derivative Filter

Neuro-Fuzzy Based ECG Signal Classification with A Gaussian Derivative Filter

Hybrid Features and Classifier for Classification of ECG Signal

In this research, we have proposed an efficient technique to classify beat from ECG database. The proposed technique is composed into three stages, 1) pre processing 2) Hybrid feature extraction 3) hybrid feature classifier. The beat signals are initially taken from the physiobank ATM and in the pre-processing stage the beat signals are made suitable for feature extraction. For efficient feature extraction we use hybrid feature extractor. The hybrid feature extraction is done in three steps, i) Morphological based feature extraction ii) Haar wavelet based feature extraction iii) Tri-spectrum based feature extraction. Once the feature is extracted the hybrid classifier is used to classify the beat signal as normal or abnormal. Beat classification studies are conducted on the MIT-BIH Arrhythmia Database using three efficient features like as morphological, wavelet and trispectrum. The beat classification system based morphological information gives an accuracy of 68%, wavelet information gives an accuracy of 78%, trispectrum information gives an accuracy of 70%, combined morphological with wavelet information gives an accuracy of 77%, combined morphological with trispectral information gives an accuracy of 70%. By combining the evidence from both the morphological, wavelet and trispectrum features, an accuracy of 91% is obtained, indicating that ECG beat information is present in the hybrid features.

Classification of ECG Signal during Atrial Fibrillation Using Autoregressive Modeling

Atrial fibrillation (AF) is a common type of arrhythmia that causes death in the adults .The Auto regressive (AR) coefficients characterize the features of AF. The AR coefficients are measured for every 15 second duration of the ECG and the features are extracted using Burg's method. These features are classified using the different statistical classifiers such as kernel Support Vector Machine (KSVM) and K- Nearest Neighbor (KNN). The performance of these classifiers is evaluated on signals obtained from MIT-BIH Atrial Fibrillation Database.The effect of AR model order and data length is tested on the classification results.

Automatic classification of slow-wave sleep and REM-sleep stages using somnographic ECG signal: some preliminary results for obese and no-obese patients

A new approach of automatic classification of slow–wave sleep and REM sleep is proposed in this paper. Our approach is based on automatic classification of polysomnographic ECG signals on sleep stages. This method is applied to obese persons (weight of more than 100 kg) and no–obese persons (weight of less than 100 kg). We show the pre–processing technique of the ECG signals and, at the same time, the identification and elimination of the different types of artefacts which are contained in the signal. The automatic classification of the slow–deep sleep and the rapid eye movement sleep called in this work REM sleep consists of extracting physiological indicators that characterise these two sleep stages through the polysomnographic ECG signal. In other words, this classification is based on the analysis of the cardiac rhythm during a night's sleep. We tested our approach on ECG signals from the PHYSIOBANK database. Our experimental results show the significance of this classification method. In fact, we arrive at a precision in the order of 92.28% in the classification of slow–wave sleep and in the order of 91.82% in the classification of REM sleep. We calculated the precision of the classification compared to the results of the experts who used four physiological signals: EEG, EOG, EMG and ECG.

Fuzzy Classification of ECG Signals Using A QRS-Like FIR Filter Bank with Lattice Structures

Fuzzy Classification of ECG Signals Using A QRS-Like FIR Filter Bank with Lattice Structures

A Novel Approach to ECG Classification Based upon Two-Layered HMMs in Body Sensor Networks

This paper presents a novel approach to ECG signal filtering and classification. Unlike the traditional techniques which aim at collecting and processing the ECG signals with the patient being still, lying in bed in hospitals, our proposed algorithm is intentionally designed for monitoring and classifying the patient's ECG signals in the free-living environment. The patients are equipped with wearable ambulatory devices the whole day, which facilitates the real-time heart attack detection. In ECG preprocessing, an integral-coefficient-band-stop (ICBS) filter is applied, which omits time-consuming floating-point computations. In addition, two-layered Hidden Markov Models (HMMs) are applied to achieve ECG feature extraction and classification. The periodic ECG waveforms are segmented into ISO intervals, P subwave, QRS complex and T subwave respectively in the first HMM layer where expert-annotation assisted Baum-Welch algorithm is utilized in HMM modeling. Then the corresponding interval features are selected and applied to categorize the ECG into normal type or abnormal type (PVC, APC) in the second HMM layer. For verifying the effectiveness of our algorithm on abnormal signal detection, we have developed an ECG body sensor network (BSN) platform, whereby real-time ECG signals are collected, transmitted, displayed and the corresponding classification outcomes are deduced and shown on the BSN screen.

On the application of optimal wavelet filter banks for ECG signal classification

This paper discusses ECG signal classification after parametrizing the ECG waveforms in the wavelet domain. Signal decomposition using perfect reconstruction quadrature mirror filter banks can provide a very parsimonious representation of ECG signals. In the current work, the filter parameters are adjusted by a numerical optimization algorithm in order to minimize a cost function associated to the filter cut-off sharpness. The goal consists of achieving a better compromise between frequency selectivity and time resolution at each decomposition level than standard orthogonal filter banks such as those of the Daubechies and Coiflet families. Our aim is to optimally decompose the signals in the wavelet domain so that they can be subsequently used as inputs for training to a neural network classifier.

New Modeling for Generation of Normal and Abnormal Heart Rate Variability Signals

This research is performed based on the modeling of biological signals. We can produce Heart Rate (HR) and Heart Rate Variability (HRV) signals synthetically using the mathematical relationships which are used as input for the Integral Pulse Frequency Modulation (IPFM) model. Previous researches were proposed same methods such as one model of ECG signal synthetically based on RBF neural network, a model based on IPFM with random threshold, method was based on the estimation of produced signals which are dependent on autonomic nervous system using IPFM model with fixed threshold, a new method based on the theory of vector space that based on time-varying uses of IPMF model (TVTIPMF) and special functions, and two different methods for producing HRV signals with controlled characteristics and structure of time-frequency (TF) for using non-stationary HRV analysis. In this paper, several chaotic maps such as Logistic Map, Henon Map, Lorenz and Tent Map have been used. Also, effects of sympathetic and parasympathetic nervous system and an internal input to the SA node and their effects in HRV signals were evaluated. In the proposed method, output amount of integrator in IPFM model was compared with chaotic threshold level. Then, final output of IPFM model was characterized as the HR and HRV signal. So, from HR and HRV signals obtaining from this model, linear features such as Mean, Median, Variance, Standard Deviation, Maximum Range, Minimum Range, Mode, Amplitude Range and frequency spectrum, and non-linear features such as Lyapunov Exponent, Shanon Entropy, log Entropy, Threshold Entropy, sure Entropy and mode Entropy were extracted from artificial HRV and compared them with characteristics as extracted from natural HRV signal. Also, in this paper two patients that called high sympathetic Balance and Cardiovascular Autonomy Neuropathy (CAN) which is detected and evaluated by HRV signals were simulated. These signals by changing the values of the some coefficients of the normal simulated signal and with extracted frequency feature from these signals were simulated. For final generation of these abnormal signals, frequency features such as energy of low frequency band (EL), energy of high frequency band (HL), ratio of energy in low frequency band to the energy in high frequency band (EL/EH), ratio of energy in low frequency band to the energy in all frequency band (EL/ET) and ratio of energy in high frequency band to the energy in all frequency band (EH/ET) from abnormal signals were extracted and compared with these extracted values from normal signals. The results were closely correlated with the real data which confirm the effectiveness of the proposed model. Various signals derived from the output of this model can be used for final analysis of the HRV signals, such as arrhythmia detection and classification of ECG and HRV signals. One of the applications of the proposed model is the easy evaluation of diagnostic ECG signal processing devices. Such a model can also be used in signal compression and telemedicine application.

ANN-based Classifier for PAF Prediction

paper we propose an artificial neural network as a based classifier for prediction of Paroxysmal Atrial Fibrillation (PAF). PAF is a really life threatening disease and it is the result of irregular and repeated depolarization of the atria. We used PAF prediction data base which include 30-min. period of 100 ECG recorded signals. We divide the 30-min preceding the PAF into 6 periods with 5-min each. In each suggested period we get the classification result using ANN. The results show that we can predict the PAF accurately in 5-min & 20-min prior the PAF. In these two periods, the measured sensitivity, specificity, positive predictivity and accuracy show better and significant results comparable to the other periods. Also the results outperform the obtained results in the same field in the literature. 1. INRODUCTION Classification of ECG signal is an important area in biomedical signal processing. Paroxysmal atrial fibrillation (PAF) of the heart muscle is defined as short duration episodes of AF lasting from 2min. to less than 7 days, while chronic AF is defined as lasting more than 7 days. The main reason for this is not the immediate effect of the onset of atrial fibrillation over the patient's health (AF detection) but the long-term effects: increase in heart muscle fatigue, increase in thromboembolic and stroke events due to the formation of blood clots and an irregular onset that makes it hard to detect on normal ECG tests. Thus it is necessary for cardiologists to benefit from a robust and precise tool that could predict the onset of such events, in order to prevent them by defibrillation, drug treatment and anti- tachycardia pacing techniques. Chronic AF is usually preceded by (PAF). Therefore, in addition to use anti-arrhythmic drugs, the physicians are trying to develop pacing devices in order to surpass the onset of AF. The automated method to predict the onset PAF is interesting topic to help treating this problem. During recent years several researchers proposed many techniques to predict the onset of PAF. Useful reviews describing different techniques for PAF or chronic AF prediction, from technical to clinical points of view (1-4). The Computers in Cardiology Challenge 2001 revealed a maximum obtained accuracy of about 80% (5-7). Hariton, et al., (8) proposed a new method for PAF automatic prediction based on heart rate variability (HRV) metrics and morphologic variability (MV), and (HRV+MV) decision rule, the obtained specificity and sensitivity are between (83.93%- 89.29%), (84.51%-89,44%) respectively. Artificial neural network (ANN) in recent years has proved to be an advanced tool in solving classification (9-11), wavelets proved usefulness in feature extraction from non-stationary signal like ECG (12-13). In general, these above prediction models are able to detect the transition to PAF events with accuracies of 70-90%, by means of records of at least tens of minutes and rather complex analysis procedures. In the present work, two set of features are extracted: Feature set-1 (FS-1), directly from ECG signal and feature set-2 (FS-2) with the aid of continuous wavelet transform(CWT), which converts the time domain signal to time-frequency domain where several features can be carefully extracted, the extracted features are then applied to ANN to classify the normal object from that one who suffers from PAF.

Similarity-Based Fuzzy Classification of ECG and Capnogram Signals

A method for ECG and capnogram signals classification is proposed based on fuzzy similarity evaluation, where shape exchange algorithm and fuzzy inference are combined. It aims to be applied to quasi-periodic biomedical signals and has low computational cost. On the experiments for atrial fibrillation (AF) classification using two databases: MIT-BIH AF and MITBIH Normal Sinus Rhythm, values of 100%, 94.4%, and 97.6% for sensitivity, specificity, and accuracy respectively, and execution time of 0.6 s are obtained. The proposal is capable of been extended to classify different diseases, from ECG and capnogram signals, such as: Brugada syndrome, AV block, hypoventilation, and asthma among others to be implemented in low computational resources devices.

A model presented for classification ECG signals base on Case-Based Reasoning

Early detection of heart diseases/abnormalities can prolong life and enhance the quality of living through appropriate treatment; thus classifying cardiac signals will be helped to immediate diagnosing of heart beat type in cardiac patients. The present paper utilizes the case base reasoning (CBR) for classification of ECG signals. Four types of ECG beats (normal beat, congestive heart failure beat, ventricular tachyarrhythmia beat and atrial fibrillation beat) obtained from the PhysioBank database was classified by the proposed CBR model. The main purpose of this article is classifying heart signals and diagnosing the type of heart beat in cardiac patients that in proposed CBR (Case Base Reasoning) system, Training and testing data for diagnosing and classifying types of heart beat have been used. The evaluation results from the model are shown that the proposed model has high accuracy in classifying heart signals and helps to clinical decisions for diagnosing the type of heart beat in cardiac patients which indeed has high impact on diagnosing the type of heart beat aided computer.

Classification of ECG Signals using ANN with Resilient Back Propagation Algorithm

Electrocardiogram is one important physiological signal, which is used in assessing cardiac health. The extraction of features used for identification of the state of ECG is discussed in this paper. Using MAT LAB programs/tools, different statistical features are extracted from both normal and arrhythmia spectra. These features include arithmetic mean, median, variance, residuals on curve fitting etc. The values of the feature vector reveal information regarding cardiac health state. Then a classical multilayer feed forward neural network with back propagation algorithm is employed to serve as a classifier of the feature vector, giving 100% successful results for the specific data set considered.

A parallel genetic algorithm for adaptive hardware and its application to ECG signal classification

This paper presents a parallel genetic algorithm (GA) called the cellular compact genetic algorithm (c-cGA) and its implementation for adaptive hardware. An adaptive hardware based on the c-cGA is proposed to automate real-time classification of ECG signals. The c-cGA not only provides a strong search capability while maintaining genetic diversity using multiple GAs but also has a cellular-like structure and is a straight-forward algorithm suitable for hardware implementation. The c-cGA hardware and an adaptive digital filter structure also perform an adaptive feature selection in real time. The c-cGA is applied to a block-based neural network (BbNN) for online learning in the hardware. Using an adaptive hardware approach based on the c-cGA, an adaptive hardware system for classifying ECG signals is feasible. The proposed adaptive hardware can be implemented in a field programmable gate array (FPGA) for an adaptive embedded system applied to personalised ECG signal classifications for long-term patient monitoring.

Classification of ECG Signals Using Extreme Learning Machine

An Electrocardiogram or ECG is an electrical recording of the heart and is used in the investigation of heart disease. This ECG can be classified as normal and abnormal signals. The classification of the ECG signals is presently performed with the support vector machine. The generalization performance of the SVM classifier is not sufficient for the correct classification of ECG signals. To overcome this problem the ELM classifier is used which works by searching for the best value of the parameters that tune its discriminant function, and upstream by looking for the best subset of features that feed the classifier. The experiments were conducted on the ECG data from the Physionet arrhythmia database to classify five kinds of abnormal waveforms and normal beats. In this paper a thorough experimental study was done to show the superiority of the generalization capability of the Extreme Learning Machine (ELM) is presented and compared with support vector machine (SVM) approach in the automatic classification of ECG beats. In particular, the sensitivity of the ELM classifier is tested and that is compared with SVM combined with two classifiers, they are the k-nearest neighbor classifier (kNN) and the radial basis function neural network classifier (RBF), with respect to the curse of dimensionality and the number of available training beats. The obtained results clearly confirm the superiority of the ELM approach as compared to traditional classifiers.

A wavelet optimization approach for ECG signal classification

Wavelets have proved particularly effective for extracting discriminative features in ECG signal classification. In this paper, we show that wavelet performances in terms of classification accuracy can be pushed further by customizing them for the considered classification task. A novel approach for generating the wavelet that best represents the ECG beats in terms of discrimination capability is proposed. It makes use of the polyphase representation of the wavelet filter bank and formulates the design problem within a particle swarm optimization (PSO) framework. Experimental results conducted on the benchmark MIT/BIH arrhythmia database with the state-of-the-art support vector machine (SVM) classifier confirm the superiority in terms of classification accuracy and stability of the proposed method over standard wavelets (i.e., Daubechies and Symlet wavelets).

Classification of electrocardiogram signals with support vector machines and extreme learning machine

An Electrocardiogram or ECG is an electrical recording of the heart and is used in the investigation of heart disease. This ECG can be classified as normal and abnormal signals. The classification of the ECG signals is presently performed with the support vector machine. The generalization performance of the SVM classifier is not sufficient for the correct classification of ECG signals. To overcome this problem, the ELM classifier is used which works by searching for the best value of the parameters that tune its discriminant function and upstream by looking for the best subset of features that feed the classifier. The experiments were conducted on the ECG data from the Physionet arrhythmia database to classify five kinds of abnormal waveforms and normal beats. In this paper, a thorough experimental study was done to show the superiority of the generalization capability of the Extreme Learning Machine (ELM) that is presented and compared with support vector machine (SVM) approach in the automatic classification of ECG beats. In particular, the sensitivity of the ELM classifier is tested and that is compared with SVM combined with two classifiers, and they are the k-nearest Neighbor Classifier and the radial basis function neural network classifier, with respect to the curse of dimensionality and the number of available training beats. The obtained results clearly confirm the superiority of the ELM approach as compared with traditional classifiers.

ECG Signal Compression and Classification Algorithm With Quad Level Vector for ECG Holter System

An ECG signal processing method with quad level vector (QLV) is proposed for the ECG holter system. The ECG processing consists of the compression flow and the classification flow, and the QLV is proposed for both flows to achieve better performance with low-computation complexity. The compression algorithm is performed by using ECG skeleton and the Huffman coding. Unit block size optimization, adaptive threshold adjustment, and 4-bit-wise Huffman coding methods are applied to reduce the processing cost while maintaining the signal quality. The heartbeat segmentation and the R-peak detection methods are employed for the classification algorithm. The performance is evaluated by using the Massachusetts Institute of Technology-Boston's Beth Israel Hospital Arrhythmia Database, and the noise robust test is also performed for the reliability of the algorithm. Its average compression ratio is 16.9:1 with 0.641% percentage root mean square difference value and the encoding rate is 6.4 kbps. The accuracy performance of the R-peak detection is 100% without noise and 95.63% at the worst case with -10-dB SNR noise. The overall processing cost is reduced by 45.3% with the proposed compression techniques.