Robust algorithm for arrhythmia classification in ECG using extreme learning machine

Evaluation of a decided sample size in machine learning applications

BACKGROUND

An appropriate sample size is essential for obtaining a precise and reliable outcome of a study. In machine learning (ML), studies with inadequate samples suffer from overfitting of data and have a lower probability of producing true effects, while the increment in sample size increases the accuracy of prediction but may not cause a significant change after a certain sample size. Existing statistical approaches using standardized mean difference, effect size, and statistical power for determining sample size are potentially biased due to miscalculations or lack of experimental details. This study aims to design criteria for evaluating sample size in ML studies. We examined the average and grand effect sizes and the performance of five ML methods using simulated datasets and three real datasets to derive the criteria for sample size. We systematically increase the sample size, starting from 16, by randomly sampling and examine the impact of sample size on classifiers' performance and both effect sizes. Tenfold cross-validation was used to quantify the accuracy.

RESULTS

The results demonstrate that the effect sizes and the classification accuracies increase while the variances in effect sizes shrink with the increment of samples when the datasets have a good discriminative power between two classes. By contrast, indeterminate datasets had poor effect sizes and classification accuracies, which did not improve by increasing sample size in both simulated and real datasets. A good dataset exhibited a significant difference in average and grand effect sizes. We derived two criteria based on the above findings to assess a decided sample size by combining the effect size and the ML accuracy. The sample size is considered suitable when it has appropriate effect sizes (≥ 0.5) and ML accuracy (≥ 80%). After an appropriate sample size, the increment in samples will not benefit as it will not significantly change the effect size and accuracy, thereby resulting in a good cost-benefit ratio.

CONCLUSION

We believe that these practical criteria can be used as a reference for both the authors and editors to evaluate whether the selected sample size is adequate for a study.

A Novel DCNN-ELM Hybrid Framework For Face Mask Detection

The Coronavirus disease (2019) has caused massive destruction of human lives and capital around the world. The latest variant Omicron is proved to be the most infectious of all its previous counterparts – Alpha, Beta and Delta. Various measures are identified, tested and implemented to minimize the attack on humans. Face masks are one of those measures that are shown to be very effective in containing the infection. However, it requires continuous monitoring for law enforcement. In the present manuscript, a detailed research investigation using different ablation studies is carried out to develop the framework for face mask recognition using pre-trained deep convolution neural networks (DCNN) models used in conjunction with a fast single layer feed-forward neural network (SLFNN) commonly known as Extreme Learning Machine (ELM) as classification technique. The ELM is well known for its real time data processing capabilities and has been successfully applied both for regression and classification problems of image processing and biomedical domain. It is for the first time that in this paper we have proposed the use of ELM as classifier for face mask detection. As a precursor to this, for feature selection, six pre-trained DCNNs such as Xception, Vgg16, Vgg19, ResNet50, ResNet 101 and ResNet152 are tested for this purpose. The best testing accuracy is obtained in case of ResNet152 transfer learning model used with ELM as the classifier. The performance evaluation through different ablation studies on testing accuracy explicitly proves that ResNet152 - ELM hybrid architecture is not only the best among the selected transfer learning models but also proves so when it is compared with several other classifiers used for the face mask detection operation. Through this investigation, novelty of the use of ResNet152 + ELM for face mask detection framework in real time domain is established.

Information fusion and multi-classifier system for miner fatigue recognition in plateau environments based on electrocardiography and electromyography signals

BACKGROUND AND OBJECTIVE

Human factors are important contributors to accidents, especially human error induced by fatigue. In this study, field tests and analyses were conducted on physiological indexes extracted from electrocardiography (ECG) and electromyography (EMG) signals in miners working under the extreme conditions of a plateau environment. To provide insights into models for fatigue classification and recognition based on machine learning, multi-modal feature information fusion and miner fatigue identification based on ECG and EMG signals as physiological indicators were studied.

METHODS

Fifty-five miners were randomly selected as field test subjects, and characteristic signals were extracted from 110 groups of ECG and EMG signals as the basic signals for fatigue analysis. We conducted principal component analysis (PCA) and grey relational analysis (GRA) on the measurement indicators. Support vector machine (SVM), random forest (RF) and extreme gradient boosting (XG-Boost) machine learning models were used for fatigue classification based on multi-modal information fusion. The area under the receiver operating characteristic (ROC) curve and the confusion matrix were used to evaluate the performance of the recognition models.

RESULTS

The ECG and EMG signals showed obvious changes with fatigue. The results of fatigue model identification showed that PCA feature fusion was superior to GRA feature fusion for all three machine learning approaches, and XG-Boost achieved the best performance, with a recognition accuracy of 89.47%, a sensitivity and specificity of 100%, and an AUC of 0.90. The SVM model also showed good recognition performance (89.47% accuracy, AUC=0.89). The worst performance was that of the RF model, with a recognition accuracy of only 78.95%.

CONCLUSIONS

This study shows that the physiological indexes of ECG and EMG exhibit obvious, regular changes with fatigue and that it is feasible to use SVM, RF and XG-Boost models for miner fatigue identification. The PCA fusion technique can improve the identification accuracy more than the GRA method. XG-Boost classification yields the best accuracy and robustness. This study can serve as a reference for clinical research on the identification of human fatigue at high altitudes and for the clinical study of acute mountain sickness and human acclimatization to high altitudes.

Impact of Data Transformation: An ECG Heartbeat Classification Approach

Cardiovascular diseases continue to be a significant global health threat. The electrocardiogram (ECG) signal is a physiological signal that plays a major role in preventing severe and even fatal heart diseases. The purpose of this research is to explore a simple mathematical feature transformation that could be applied to ECG signal segments in order to improve the detection accuracy of heartbeats, which could facilitate automated heart disease diagnosis. Six different mathematical transformation methods were examined and analyzed using 10s-length ECG segments, which showed that a reciprocal transformation results in consistently better classification performance for normal vs. atrial fibrillation beats and normal vs. atrial premature beats, when compared to untransformed features. The second best data transformation in terms of heartbeat detection accuracy was the cubic transformation. Results showed that applying the logarithmic transformation, which is considered the go-to data transformation, was not optimal among the six data transformations. Using the optimal data transformation, the reciprocal, can lead to a 35.6% accuracy improvement. According to the overall comparison tested by different feature engineering methods, classifiers, and different dataset sizes, performance improvement also reached 4.7%. Therefore, adding a simple data transformation step, such as the reciprocal or cubic, to the extracted features can improve current automated heartbeat classification in a timely manner.

Automated detection of arrhythmia from electrocardiogram signal based on new convolutional encoded features with bidirectional long short-term memory network classifier

Early detection of cardiac arrhythmia is needed to reduce mortality. Automatically detecting the cardiac arrhythmias is a very challenging task. In this paper, a new deep convolutional encoded feature (CEF) based on non-linear compression composition is applied to diminish the ECG signal segment size. Bidirectional long short-term memory (BLSTM) network classifier has been proposed to detect arrhythmias from the ECG signal, which is encoded by the convolutional encoder. These encoded features are used as the input to BLSTM network classifier. For performance comparison, three other classifiers, namely unidirectional long short-term memory (ULSTM) network, gated recurrent Unit (GRU) and multilayer perceptron, are designed. The experimental studies detect and classify arrhythmias present in the MIT-BIH arrhythmia database into five different heartbeat classes. These heartbeat classes are normal (N), left bundle branch block (L), right bundle branch block(R), paced (P) and premature ventricular contraction (V). Evaluation of performance and system efficiency has been done with the help of four different types of evaluation criteria which are overall accuracy, precision, recall, and F-score. The experimental results indicate that the BLSTM network has achieved an overall accuracy of 99.52% with the processing time of only 6.043 s.

Classification of normal sinus rhythm, abnormal arrhythmia and congestive heart failure ECG signals using LSTM and hybrid CNN-SVM deep neural networks

Effective monitoring of heart patients according to heart signals can save a huge amount of life. In the last decade, the classification and prediction of heart diseases according to ECG signals has gained great importance for patients and doctors. In this paper, the deep learning architecture with high accuracy and popularity has been proposed in recent years for the classification of Normal Sinus Rhythm, (NSR) Abnormal Arrhythmia (ARR) and Congestive Heart Failure (CHF) ECG signals. The proposed architecture is based on Hybrid Alexnet-SVM (Support Vector Machine). 96 Arrhythmia, 30 CHF, 36 NSR signals are available in a total of 192 ECG signals. In order to demonstrate the classification performance of deep learning architectures, ARR, CHR and NSR signals are firstly classified by SVM, KNN algorithm, achieving 68.75% and 65.63% accuracy. The signals are then classified in their raw form with LSTM (Long Short Time Memory) with 90.67% accuracy. By obtaining the spectrograms of the signals, Hybrid Alexnet-SVM algorithm is applied to the images and 96.77% accuracy is obtained. The results show that with the proposed deep learning architecture, it classifies ECG signals with higher accuracy than conventional machine learning classifiers.

Frobenius and nuclear hybrid norm penalized robust principal component analysis for transient impulsive feature detection of rolling bearings

Transient impulsive feature detection is of vital importance in fault diagnosis of rolling bearing. However, the transient impulsive feature of rolling bearing is always heavily buried in the noise contaminated signal, which makes it difficult to be detected. Robust principal component analysis (PRCA) is an effective approach to exploit the underlying structure from the corrupted observation, where the decomposed low-rank matrix (LRM) and the sparse matrix can represent the useful diagnostic information and the unwanted background noise respectively. In this study, a Frobenius and nuclear hybrid norm penalized RPCA (FNHN-RPCA) is served as a specific RPCA solver on account of it has a great ability to approach to the rank of the LRM and make the execution procedure efficiently To make the recorded signal suitable for the input criterion of the RPCA solver, a fault information matrix (FIM) construction method is proposed to arrange the recorded signal into a matrix form. After the RPCA solver is conducted on it, a reversed recovery operation is also proposed to rearrange the two dimensional LRM into a one-dimensional signal form. To confirm all recorded data is processed by the RPCA solver, both the forward and backward FIMs are constructed and a synthesis of the recovered signals from both the forward and backward FIMs is served as the final transient impulsive feature enhanced signal. The diagnostic results on simulated and experimental case studies verify that the presented technique is suitable for transient impulsive feature detection of rolling bearings even when the test bearing works in a low speed operating condition or the operating environment is in the presence of random impact interference.

Computational approaches for detection of cardiac rhythm abnormalities: Are we there yet?

The analysis of an electrocardiogram (ECG) is able to provide vital information on the electrical activity of the heart and is crucial for the accurate diagnosis of cardiac arrhythmias. Due to the nature of some arrhythmias, this might be a time-consuming and difficult to accomplish process. The advent of novel machine learning technologies in this field has a potential to revolutionise the use of the ECG. In this review, we outline key advances in ECG analysis for atrial, ventricular and complex multiform arrhythmias, as well as discuss the current limitations of the technology and the barriers that must be overcome before clinical integration is feasible.

A novel ECG signal classification method using DEA-ELM

Electrocardiogram (ECG) signals represent the electrical mobility of the human heart. In recent years, computer-aided systems have helped to cardiologists in the detection, classification and diagnosis of ECG. The aim of this paper is to optimize the number hidden neurons of the traditional Extreme Learning Machine (ELM) using Differential Evolution Algorithm (DEA) and contribute to the classification of ECG signals with a higher accuracy rate. In this paper, publicly ECG records in Physionet was utilized. Pan-Tompkins technique (PTT) and Discrete Wavelet Transform (DWT) approaches were implemented to obtain characteristic properties which are PR period, QT period, ST period and QRS wave of ECG signals. Then, ELM was executed to the ECG samples. Lastly, DEA on software ELM was developed for the assign of the number of hidden neurons, which were used in the ELM algorithm. The performance criterions were used in order to compare the performance of the classification exerted. Concordantly, it was realized that the highest classification achievement values were reached to Accuracy 97.5% and values 93 of number of hidden neurons, with the practice improved with the DEA compared to conventional ELM.

A deep neural network learning algorithm outperforms a conventional algorithm for emergency department electrocardiogram interpretation

BACKGROUND

Cardiologs® has developed the first electrocardiogram (ECG) algorithm that uses a deep neural network (DNN) for full 12‑lead ECG analysis, including rhythm, QRS and ST-T-U waves. We compared the accuracy of the first version of Cardiologs® DNN algorithm to the Mortara/Veritas® conventional algorithm in emergency department (ED) ECGs.

METHODS

Individual ECG diagnoses were prospectively mapped to one of 16 pre-specified groups of ECG diagnoses, which were further classified as "major" ECG abnormality or not. Automated interpretations were compared to blinded experts'. The primary outcome was the performance of the algorithms in finding at least one "major" abnormality. The secondary outcome was the proportion of all ECGs for which all groups were identified, with no false negative or false positive groups ("accurate ECG interpretation"). Additionally, we measured sensitivity and positive predictive value (PPV) for any abnormal group.

RESULTS

Cardiologs® vs. Veritas® accuracy for finding a major abnormality was 92.2% vs. 87.2% (p < 0.0001), with comparable sensitivity (88.7% vs. 92.0%, p = 0.086), improved specificity (94.0% vs. 84.7%, p < 0.0001) and improved positive predictive value (PPV 88.2% vs. 75.4%, p < 0.0001). Cardiologs® had accurate ECG interpretation for 72.0% (95% CI: 69.6-74.2) of ECGs vs. 59.8% (57.3-62.3) for Veritas® (P < 0.0001). Sensitivity for any abnormal group for Cardiologs® and Veritas®, respectively, was 69.6% (95CI 66.7-72.3) vs. 68.3% (95CI 65.3-71.1) (NS). Positive Predictive Value was 74.0% (71.1-76.7) for Cardiologs® vs. 56.5% (53.7-59.3) for Veritas® (P < 0.0001).

CONCLUSION

Cardiologs' DNN was more accurate and specific in identifying ECGs with at least one major abnormal group. It had a significantly higher rate of accurate ECG interpretation, with similar sensitivity and higher PPV.

Novel Scalable Deep Learning Approaches for Big Data Analytics Applied to ECG Processing

Big data analytics and deep learning are nowadays two of the most active research areas in computer science. As the data is becoming bigger and bigger, deep learning has a very important role to play in data analytics, and big data technologies will give it huge opportunities for different sectors. Deep learning brings new challenges especially when it comes to large amounts of data, the volume of datasets has to be processed and managed, also data in various applications come in a streaming way and deep learning approaches have to deal with this kind of applications. In this paper, the authors propose two novel approaches for discriminative deep learning, namely LS-DSN, and StreamDSN that are inspired from the deep stacking network algorithm. Two versions of the gradient descent algorithm were used to train the proposed algorithms. The experiment results have shown that the algorithms gave satisfying accuracy results and scale well when the size of data increases. In addition, StreamDSN algorithm have been applied to classify beats of ECG signals and provided good promising results.

A Spatial Pyramid Pooling-Based Deep Convolutional Neural Network for the Classification of Electrocardiogram Beats

An accurate electrocardiogram (ECG) beat classification can benefit the diagnosis of the cardiovascular disease. Deep convolutional neural networks (CNN) can automatically extract valid features from data, which is an effective way for the classification of the ECG beats. However, the fully-connected layer in CNNs requires a fixed input dimension, which limits the CNNs to receive fixed-scale inputs. Signals of different scales are generally processed into the same size by segmentation and downsampling. If information loss occurs during a uniformly-sized process, the classification accuracy will ultimately be affected. To solve this problem, this paper constructs a new CNN framework spatial pyramid pooling (SPP) method, which solves the deficiency caused by the size of input data. The Massachusetts Institute of Technology-Biotechnology (MIT-BIH) arrhythmia database is employed as the training and testing data for the classification of heartbeat signals into six categories. Compared with the traditional method, which may lose a large amount of important information and easy to be over-fitted, the robustness of the proposed method can be guaranteed by extracting data features from different sizes. Experimental results show that the proposed architecture network can extract more high-quality features and exhibits higher classification accuracy (94%) than the traditional deep CNNs (90.4%).

Spatial extreme learning machines: An application on prediction of disease counts

Extreme learning machines have gained a lot of attention by the machine learning community because of its interesting properties and computational advantages. With the increase in collection of information nowadays, many sources of data have missing information making statistical analysis harder or unfeasible. In this paper, we present a new model, coined spatial extreme learning machine, that combine spatial modeling with extreme learning machines keeping the nice properties of both methodologies and making it very flexible and robust. As explained throughout the text, the spatial extreme learning machines have many advantages in comparison with the traditional extreme learning machines. By a simulation study and a real data analysis we present how the spatial extreme learning machine can be used to improve imputation of missing data and uncertainty prediction estimation.

Fusion of ECG and ABP signals based on wavelet transform for cardiac arrhythmias classification

BACKGROUND AND OBJECTIVE

Each of Electrocardiogram (ECG) and Atrial Blood Pressure (ABP) signals contain information of cardiac status. This information can be used for diagnosis and monitoring of diseases. The majority of previously proposed methods rely only on ECG signal to classify heart rhythms. In this paper, ECG and ABP were used to classify five different types of heart rhythms. To this end, two mentioned signals (ECG and ABP) have been fused.

METHODS

These physiological signals have been used from MINIC physioNet database. ECG and ABP signals have been fused together on the basis of the proposed Discrete Wavelet Transformation fusion technique. Then, some frequency features were extracted from the fused signal. To classify the different types of cardiac arrhythmias, these features were given to a multi-layer perceptron neural network.

RESULTS

In this study, the best results for the proposed fusion algorithm were obtained. In this case, the accuracy rates of 96.6%, 96.9%, 95.6% and 93.9% were achieved for two, three, four and five classes, respectively. However, the maximum classification rate of 89% was obtained for two classes on the basis of ECG features.

CONCLUSIONS

It has been found that the higher accuracy rates were acquired by using the proposed fusion technique. The results confirmed the importance of fusing features from different physiological signals to gain more accurate assessments.

Artificial Intelligence versus Doctors' Intelligence: A Glance on Machine Learning Benefaction in Electrocardiography

Computational machine learning, especially self-enhancing algorithms, prove remarkable effectiveness in applications, including cardiovascular medicine. This review summarizes and cross-compares the current machine learning algorithms applied to electrocardiogram interpretation. In practice, continuous real-time monitoring of electrocardiograms is still difficult to realize. Furthermore, automated ECG interpretation by implementing specific artificial intelligence algorithms is even more challenging. By collecting large datasets from one individual, computational approaches can assure an efficient personalized treatment strategy, such as a correct prediction on patient-specific disease progression, therapeutic success rate and limitations of certain interventions, thus reducing the hospitalization costs and physicians’ workload. Clearly such aims can be achieved by a perfect symbiosis of a multidisciplinary team involving clinicians, researchers and computer scientists. Summarizing, continuous cross-examination between machine intelligence and human intelligence is a combination of precision, rationale and high-throughput scientific engine integrated into a challenging framework of big data science.

Systematic mapping study of data mining–based empirical studies in cardiology

Data mining provides the methodology and technology to transform huge amount of data into useful information for decision making. It is a powerful process to extract knowledge and discover new patterns embedded in large data sets. Data mining has been increasingly used in medicine, particularly in cardiology. In fact, data mining applications can greatly benefits all parts involved in cardiology such as patients, cardiologists and nurses. This article aims to perform a systematic mapping study so as to analyze and synthesize empirical studies on the application of data mining techniques in cardiology. A total of 142 articles published between 2000 and 2015 were therefore selected, studied and analyzed according to the four following criteria: year and channel of publication, research type, medical task and empirical type. The results of this mapping study are discussed and a list of recommendations for researchers and cardiologists is provided.

Neuromorphic Hardware Architecture Using the Neural Engineering Framework for Pattern Recognition

We present a hardware architecture that uses the neural engineering framework (NEF) to implement large-scale neural networks on field programmable gate arrays (FPGAs) for performing massively parallel real-time pattern recognition. NEF is a framework that is capable of synthesising large-scale cognitive systems from subnetworks and we have previously presented an FPGA implementation of the NEF that successfully performs nonlinear mathematical computations. That work was developed based on a compact digital neural core, which consists of 64 neurons that are instantiated by a single physical neuron using a time-multiplexing approach. We have now scaled this approach up to build a pattern recognition system by combining identical neural cores together. As a proof of concept, we have developed a handwritten digit recognition system using the MNIST database and achieved a recognition rate of 96.55%. The system is implemented on a state-of-the-art FPGA and can process 5.12 million digits per second. The architecture and hardware optimisations presented offer high-speed and resource-efficient means for performing high-speed, neuromorphic, and massively parallel pattern recognition and classification tasks.

Vowel Imagery Decoding toward Silent Speech BCI Using Extreme Learning Machine with Electroencephalogram

The purpose of this study is to classify EEG data on imagined speech in a single trial. We recorded EEG data while five subjects imagined different vowels, /a/, /e/, /i/, /o/, and /u/. We divided each single trial dataset into thirty segments and extracted features (mean, variance, standard deviation, and skewness) from all segments. To reduce the dimension of the feature vector, we applied a feature selection algorithm based on the sparse regression model. These features were classified using a support vector machine with a radial basis function kernel, an extreme learning machine, and two variants of an extreme learning machine with different kernels. Because each single trial consisted of thirty segments, our algorithm decided the label of the single trial by selecting the most frequent output among the outputs of the thirty segments. As a result, we observed that the extreme learning machine and its variants achieved better classification rates than the support vector machine with a radial basis function kernel and linear discrimination analysis. Thus, our results suggested that EEG responses to imagined speech could be successfully classified in a single trial using an extreme learning machine with a radial basis function and linear kernel. This study with classification of imagined speech might contribute to the development of silent speech BCI systems.

Semi-supervised Stacked Label Consistent Autoencoder for Reconstruction and Analysis of Biomedical Signals

OBJECTIVE

An autoencoder-based framework that simultaneously reconstruct and classify biomedical signals is proposed. Previous work has treated reconstruction and classification as separate problems. This is the first study that proposes a combined framework to address the issue in a holistic fashion.

METHODS

For telemonitoring purposes, reconstruction techniques of biomedical signals are largely based on compressed sensing (CS); these are "designed" techniques where the reconstruction formulation is based on some "assumption" regarding the signal. In this study, we propose a new paradigm for reconstruction-the reconstruction is "learned," using an autoencoder; it does not require any assumption regarding the signal as long as there is sufficiently large training data. But since the final goal is to analyze/classify the signal, the system can also learn a linear classification map that is added inside the autoencoder. The ensuing optimization problem is solved using the Split Bregman technique.

RESULTS

Experiments were carried out on reconstructing and classifying electrocardiogram (ECG) (arrhythmia classification) and EEG (seizure classification) signals.

CONCLUSION

Our proposed tool is capable of operating in a semi-supervised fashion. We show that our proposed method is better in reconstruction and more than an order magnitude faster than CS based methods; it is capable of real-time operation. Our method also yields better results than recently proposed classification methods.

SIGNIFICANCE

This is the first study offering an alternative to CS-based reconstruction. It also shows that the representation learning approach can yield better results than traditional methods that use hand-crafted features for signal analysis.

Arrhythmia recognition and classification using combined linear and nonlinear features of ECG signals

Arrhythmia is a cardiac condition caused by abnormal electrical activity of the heart, and an electrocardiogram (ECG) is the non-invasive method used to detect arrhythmias or heart abnormalities. Due to the presence of noise, the non-stationary nature of the ECG signal (i.e. the changing morphology of the ECG signal with respect to time) and the irregularity of the heartbeat, physicians face difficulties in the diagnosis of arrhythmias. The computer-aided analysis of ECG results assists physicians to detect cardiovascular diseases. The development of many existing arrhythmia systems has depended on the findings from linear experiments on ECG data which achieve high performance on noise-free data. However, nonlinear experiments characterize the ECG signal more effectively sense, extract hidden information in the ECG signal, and achieve good performance under noisy conditions. This paper investigates the representation ability of linear and nonlinear features and proposes a combination of such features in order to improve the classification of ECG data. In this study, five types of beat classes of arrhythmia as recommended by the Association for Advancement of Medical Instrumentation are analyzed: non-ectopic beats (N), supra-ventricular ectopic beats (S), ventricular ectopic beats (V), fusion beats (F) and unclassifiable and paced beats (U). The characterization ability of nonlinear features such as high order statistics and cumulants and nonlinear feature reduction methods such as independent component analysis are combined with linear features, namely, the principal component analysis of discrete wavelet transform coefficients. The features are tested for their ability to differentiate different classes of data using different classifiers, namely, the support vector machine and neural network methods with tenfold cross-validation. Our proposed method is able to classify the N, S, V, F and U arrhythmia classes with high accuracy (98.91%) using a combined support vector machine and radial basis function method.

ECG-based heartbeat classification for arrhythmia detection: A survey

An electrocardiogram (ECG) measures the electric activity of the heart and has been widely used for detecting heart diseases due to its simplicity and non-invasive nature. By analyzing the electrical signal of each heartbeat, i.e., the combination of action impulse waveforms produced by different specialized cardiac tissues found in the heart, it is possible to detect some of its abnormalities. In the last decades, several works were developed to produce automatic ECG-based heartbeat classification methods. In this work, we survey the current state-of-the-art methods of ECG-based automated abnormalities heartbeat classification by presenting the ECG signal preprocessing, the heartbeat segmentation techniques, the feature description methods and the learning algorithms used. In addition, we describe some of the databases used for evaluation of methods indicated by a well-known standard developed by the Association for the Advancement of Medical Instrumentation (AAMI) and described in ANSI/AAMI EC57:1998/(R)2008 (ANSI/AAMI, 2008). Finally, we discuss limitations and drawbacks of the methods in the literature presenting concluding remarks and future challenges, and also we propose an evaluation process workflow to guide authors in future works.

Simple and Robust Realtime QRS Detection Algorithm Based on Spatiotemporal Characteristic of the QRS Complex

The purpose of this research is to develop an intuitive and robust realtime QRS detection algorithm based on the physiological characteristics of the electrocardiogram waveform. The proposed algorithm finds the QRS complex based on the dual criteria of the amplitude and duration of QRS complex. It consists of simple operations, such as a finite impulse response filter, differentiation or thresholding without complex and computational operations like a wavelet transformation. The QRS detection performance is evaluated by using both an MIT-BIH arrhythmia database and an AHA ECG database (a total of 435,700 beats). The sensitivity (SE) and positive predictivity value (PPV) were 99.85% and 99.86%, respectively. According to the database, the SE and PPV were 99.90% and 99.91% in the MIT-BIH database and 99.84% and 99.84% in the AHA database, respectively. The result of the noisy environment test using record 119 from the MIT-BIH database indicated that the proposed method was scarcely affected by noise above 5 dB SNR (SE = 100%, PPV > 98%) without the need for an additional de-noising or back searching process.

Comparison of Support-Vector Machine and Sparse Representation Using a Modified Rule-Based Method for Automated Myocardial Ischemia Detection

An automatic method is presented for detecting myocardial ischemia, which can be considered as the early symptom of acute coronary events. Myocardial ischemia commonly manifests as ST- and T-wave changes on ECG signals. The methods in this study are proposed to detect abnormal ECG beats using knowledge-based features and classification methods. A novel classification method, sparse representation-based classification (SRC), is involved to improve the performance of the existing algorithms. A comparison was made between two classification methods, SRC and support-vector machine (SVM), using rule-based vectors as input feature space. The two methods are proposed with quantitative evaluation to validate their performances. The results of SRC method encompassed with rule-based features demonstrate higher sensitivity than that of SVM. However, the specificity and precision are a trade-off. Moreover, SRC method is less dependent on the selection of rule-based features and can achieve high performance using fewer features. The overall performances of the two methods proposed in this study are better than the previous methods.

Functional connectivity changes detected with magnetoencephalography after mild traumatic brain injury

Mild traumatic brain injury (mTBI) may affect normal cognition and behavior by disrupting the functional connectivity networks that mediate efficient communication among brain regions. In this study, we analyzed brain connectivity profiles from resting state Magnetoencephalographic (MEG) recordings obtained from 31 mTBI patients and 55 normal controls. We used phase-locking value estimates to compute functional connectivity graphs to quantify frequency-specific couplings between sensors at various frequency bands. Overall, normal controls showed a dense network of strong local connections and a limited number of long-range connections that accounted for approximately 20% of all connections, whereas mTBI patients showed networks characterized by weak local connections and strong long-range connections that accounted for more than 60% of all connections. Comparison of the two distinct general patterns at different frequencies using a tensor representation for the connectivity graphs and tensor subspace analysis for optimal feature extraction showed that mTBI patients could be separated from normal controls with 100% classification accuracy in the alpha band. These encouraging findings support the hypothesis that MEG-based functional connectivity patterns may be used as biomarkers that can provide more accurate diagnoses, help guide treatment, and monitor effectiveness of intervention in mTBI.

•

We analyzed resting state connectivity profiles in 31 mTBI patients and 55 controls.

•

We quantified frequency-specific connectivity couplings using phase-locking values.

•

Normal control networks showed dense local and sparse long-range connections.

•

TBI patient networks showed sparse local and dense long-range connections.

•

Tensor subspace analysis could classify subjects with 100% accuracy in the α band

Trends in extreme learning machines: a review

Extreme learning machine (ELM) has gained increasing interest from various research fields recently. In this review, we aim to report the current state of the theoretical research and practical advances on this subject. We first give an overview of ELM from the theoretical perspective, including the interpolation theory, universal approximation capability, and generalization ability. Then we focus on the various improvements made to ELM which further improve its stability, sparsity and accuracy under general or specific conditions. Apart from classification and regression, ELM has recently been extended for clustering, feature selection, representational learning and many other learning tasks. These newly emerging algorithms greatly expand the applications of ELM. From implementation aspect, hardware implementation and parallel computation techniques have substantially sped up the training of ELM, making it feasible for big data processing and real-time reasoning. Due to its remarkable efficiency, simplicity, and impressive generalization performance, ELM have been applied in a variety of domains, such as biomedical engineering, computer vision, system identification, and control and robotics. In this review, we try to provide a comprehensive view of these advances in ELM together with its future perspectives.

Electrocardiogram pattern recognition and analysis based on artificial neural networks and support vector machines: a review

Computer systems for Electrocardiogram (ECG) analysis support the clinician in tedious tasks (e.g., Holter ECG monitored in Intensive Care Units) or in prompt detection of dangerous events (e.g., ventricular fibrillation). Together with clinical applications (arrhythmia detection and heart rate variability analysis), ECG is currently being investigated in biometrics (human identification), an emerging area receiving increasing attention. Methodologies for clinical applications can have both differences and similarities with respect to biometrics. This paper reviews methods of ECG processing from a pattern recognition perspective. In particular, we focus on features commonly used for heartbeat classification. Considering the vast literature in the field and the limited space of this review, we dedicated a detailed discussion only to a few classifiers (Artificial Neural Networks and Support Vector Machines) because of their popularity; however, other techniques such as Hidden Markov Models and Kalman Filtering will be also mentioned.

Arrhythmia detection in single-lead ECG by combining beat and rhythm-level information

In this paper, we propose a method for detecting arrhythmia in single-lead electro-cardiogram (ECG) signal. By applying a sequence of pre-processing steps (filtering, baseline correction), beat classification and rhythm identification, six different beat-types and four abnormal rhythms are detected. Beat classification uses fast Fourier transform (FFT) as the feature and a support vector machine (SVM) classifier. Subsequently rhythm identification uses a deterministic finite state machine to detect abnormal rhythms. We evaluate the performance of our technique on the MIT-BIH database, to obtain 97% beat classification accuracy and perfect rhythm identification result.

Real-time feature extraction of P300 component using adaptive nonlinear principal component analysis

BACKGROUND

The electroencephalography (EEG) signals are known to involve the firings of neurons in the brain. The P300 wave is a high potential caused by an event-related stimulus. The detection of P300s included in the measured EEG signals is widely investigated. The difficulties in detecting them are that they are mixed with other signals generated over a large brain area and their amplitudes are very small due to the distance and resistivity differences in their transmittance.

METHODS

A novel real-time feature extraction method for detecting P300 waves by combining an adaptive nonlinear principal component analysis (ANPCA) and a multilayer neural network is proposed. The measured EEG signals are first filtered using a sixth-order band-pass filter with cut-off frequencies of 1 Hz and 12 Hz. The proposed ANPCA scheme consists of four steps: pre-separation, whitening, separation, and estimation. In the experiment, four different inter-stimulus intervals (ISIs) are utilized: 325 ms, 350 ms, 375 ms, and 400 ms.

RESULTS

The developed multi-stage principal component analysis method applied at the pre-separation step has reduced the external noises and artifacts significantly. The introduced adaptive law in the whitening step has made the subsequent algorithm in the separation step to converge fast. The separation performance index has varied from -20 dB to -33 dB due to randomness of source signals. The robustness of the ANPCA against background noises has been evaluated by comparing the separation performance indices of the ANPCA with four algorithms (NPCA, NSS-JD, JADE, and SOBI), in which the ANPCA algorithm demonstrated the shortest iteration time with performance index about 0.03. Upon this, it is asserted that the ANPCA algorithm successfully separates mixed source signals.

CONCLUSIONS

The independent components produced from the observed data using the proposed method illustrated that the extracted signals were clearly the P300 components elicited by task-related stimuli. The experiment using 350 ms ISI showed the best performance. Since the proposed method does not use down-sampling and averaging, it can be used as a viable tool for real-time clinical applications.

An arrhythmia classification algorithm using a dedicated wavelet adapted to different subjects

BACKGROUND

Numerous studies have been conducted regarding a heartbeat classification algorithm over the past several decades. However, many algorithms have also been studied to acquire robust performance, as biosignals have a large amount of variation among individuals. Various methods have been proposed to reduce the differences coming from personal characteristics, but these expand the differences caused by arrhythmia.

METHODS

In this paper, an arrhythmia classification algorithm using a dedicated wavelet adapted to individual subjects is proposed. We reduced the performance variation using dedicated wavelets, as in the ECG morphologies of the subjects. The proposed algorithm utilizes morphological filtering and a continuous wavelet transform with a dedicated wavelet. A principal component analysis and linear discriminant analysis were utilized to compress the morphological data transformed by the dedicated wavelets. An extreme learning machine was used as a classifier in the proposed algorithm.

RESULTS

A performance evaluation was conducted with the MIT-BIH arrhythmia database. The results showed a high sensitivity of 97.51%, specificity of 85.07%, accuracy of 97.94%, and a positive predictive value of 97.26%.

CONCLUSIONS

The proposed algorithm achieves better accuracy than other state-of-the-art algorithms with no intrasubject between the training and evaluation datasets. And it significantly reduces the amount of intervention needed by physicians.

BELM: Bayesian extreme learning machine

The theory of extreme learning machine (ELM) has become very popular on the last few years. ELM is a new approach for learning the parameters of the hidden layers of a multilayer neural network (as the multilayer perceptron or the radial basis function neural network). Its main advantage is the lower computational cost, which is especially relevant when dealing with many patterns defined in a high-dimensional space. This brief proposes a bayesian approach to ELM, which presents some advantages over other approaches: it allows the introduction of a priori knowledge; obtains the confidence intervals (CIs) without the need of applying methods that are computationally intensive, e.g., bootstrap; and presents high generalization capabilities. Bayesian ELM is benchmarked against classical ELM in several artificial and real datasets that are widely used for the evaluation of machine learning algorithms. Achieved results show that the proposed approach produces a competitive accuracy with some additional advantages, namely, automatic production of CIs, reduction of probability of model overfitting, and use of a priori knowledge.