The PhysioNet / Computers in Cardiology Challenge 2008: T-Wave Alternans

A systematic review of automated prediction of sudden cardiac death using ECG signals

Background. Sudden cardiac death (SCD) stands as a life-threatening cardiac event capable of swiftly claiming lives. It ranks prominently among the leading causes of global mortality, contributing to approximately 10% of deaths worldwide. The timely anticipation of SCD holds the promise of immediate life-saving interventions, such as cardiopulmonary resuscitation. However, recent strides in the realms of deep learning (DL), machine learning (ML), and artificial intelligence have ushered in fresh opportunities for the automation of SCD prediction using physiological signals. Researchers have devised numerous models to automatically predict SCD using a combination of diverse feature extraction techniques and classifiers. Methods: We conducted a thorough review of research publications ranging from 2011 to 2023, with a specific focus on the automated prediction of SCD. Traditionally, specialists utilize molecular biomarkers, symptoms, and 12-lead ECG recordings for SCD prediction. However, continuous patient monitoring by experts is impractical, and only a fraction of patients seeks help after experiencing symptoms. However, over the past two decades, ML techniques have emerged and evolved for this purpose. Importantly, since 2021, the studies we have scrutinized delve into a diverse array of ML and DL algorithms, encompassing K-nearest neighbors, support vector machines, decision trees, random forest, Naive Bayes, and convolutional neural networks as classifiers.Results. This literature review presents a comprehensive analysis of ML and DL models employed in predicting SCD. The analysis provided valuable information on the fundamental structure of cardiac fatalities, extracting relevant characteristics from electrocardiogram (ECG) and heart rate variability (HRV) signals, using databases, and evaluating classifier performance. The review offers a succinct yet thorough examination of automated SCD prediction methodologies, emphasizing current constraints and underscoring the necessity for further advancements. It serves as a valuable resource, providing valuable insights and outlining potential research directions for aspiring scholars in the domain of SCD prediction.Conclusions. In recent years, researchers have made substantial strides in the prediction of SCD by leveraging openly accessible databases such as the MIT-BIH SCD Holter and Normal Sinus Rhythm, which contains extensive 24 h recordings of SCD patients. These sophisticated methodologies have previously demonstrated the potential to achieve remarkable accuracy, reaching levels as high as 97%, and can forecast SCD events with a lead time of 30-70 min. Despite these promising outcomes, the quest for even greater accuracy and reliability persists. ML and DL methodologies have shown great promise, their performance is intrinsically linked to the volume of training data available. Most predictive models rely on small-scale databases, raising concerns about their applicability in real-world scenarios. Furthermore, these models predominantly utilize ECG and HRV signals, often overlooking the potential contributions of other physiological signals. Developing real-time, clinically applicable models also represents a critical avenue for further exploration in this field.

Standardized Database of 12-Lead Electrocardiograms with a Common Standard for the Promotion of Cardiovascular Research: KURIAS-ECG

OBJECTIVES

Electrocardiography (ECG)-based diagnosis by experts cannot maintain uniform quality because individual differences may occur. Previous public databases can be used for clinical studies, but there is no common standard that would allow databases to be combined. For this reason, it is difficult to conduct research that derives results by combining databases. Recent commercial ECG machines offer diagnoses similar to those of a physician. Therefore, the purpose of this study was to construct a standardized ECG database using computerized diagnoses.

METHODS

The constructed database was standardized using Systematized Nomenclature of Medicine Clinical Terms (SNOMED CT) and Observational Medical Outcomes Partnership-common data model (OMOP-CDM), and data were then categorized into 10 groups based on the Minnesota classification. In addition, to extract high-quality waveforms, poor-quality ECGs were removed, and database bias was minimized by extracting at least 2,000 cases for each group. To check database quality, the difference in baseline displacement according to whether poor ECGs were removed was analyzed, and the usefulness of the database was verified with seven classification models using waveforms.

RESULTS

The standardized KURIAS-ECG database consists of high-quality ECGs from 13,862 patients, with about 20,000 data points, making it possible to obtain more than 2,000 for each Minnesota classification. An artificial intelligence classification model using the data extracted through SNOMED-CT showed an average accuracy of 88.03%.

CONCLUSIONS

The KURIAS-ECG database contains standardized ECG data extracted from various machines. The proposed protocol should promote cardiovascular disease research using big data and artificial intelligence.

Machine Learning approach for TWA detection relying on ensemble data design

BACKGROUND AND OBJECTIVE

T-wave alternans (TWA) is a fluctuation of the ST-T complex of the surface electrocardiogram (ECG) on an every-other-beat basis. It has been shown to be clinically helpful for sudden cardiac death stratification, though the lack of a gold standard to benchmark detection methods limits its application and impairs the development of alternative techniques. In this work, a novel approach based on machine learning for TWA detection is proposed. Additionally, a complete experimental setup is presented for TWA detection methods benchmarking.

METHODS

The proposed experimental setup is based on the use of open-source databases to enable experiment replication and the use of real ECG signals with added TWA episodes. Also, intra-patient overfitting and class imbalance have been carefully avoided. The Spectral Method (SM), the Modified Moving Average Method (MMA), and the Time Domain Method (TM) are used to obtain input features to the Machine Learning (ML) algorithms, namely, K Nearest Neighbor, Decision Trees, Random Forest, Support Vector Machine and Multi-Layer Perceptron.

RESULTS

There were not found large differences in the performance of the different ML algorithms. Decision Trees showed the best overall performance (accuracy 0.88 ± 0.04 , precision 0.89 ± 0.05 , Recall 0.90 ± 0.05 , F1 score 0.89 ± 0.03 ). Compared to the SM (accuracy 0.79, precision 0.93, Recall 0.64, F1 score 0.76) there was an improvement in every metric except for the precision.

CONCLUSIONS

In this work, a realistic database to test the presence of TWA using ML algorithms was assembled. The ML algorithms overall outperformed the SM used as a gold standard. Learning from data to identify alternans elicits a substantial detection growth at the expense of a small increment of the false alarm.

DAE-ConvBiLSTM: End-to-end learning single-lead electrocardiogram signal for heart abnormalities detection

BACKGROUND

The electrocardiogram (ECG) is a widely used diagnostic that observes the heart activities of patients to ascertain a heart abnormality diagnosis. The artifacts or noises are primarily associated with the problem of ECG signal processing. Conventional denoising techniques have been proposed in previous literature; however, some lacks, such as the determination of suitable wavelet basis function and threshold, can be a time-consuming process. This paper presents end-to-end learning using a denoising auto-encoder (DAE) for denoising algorithms and convolutional-bidirectional long short-term memory (ConvBiLSTM) for ECG delineation to classify ECG waveforms in terms of the PQRST-wave and isoelectric lines. The denoising reconstruction using unsupervised learning based on the encoder-decoder process can be proposed to improve the drawbacks. First, The ECG signals are reduced to a low-dimensional vector in the encoder. Second, the decoder reconstructed the signals. The last, the reconstructed signals of ECG can be processed to ConvBiLSTM. The proposed architecture of DAE-ConvBiLSTM is the end-to-end diagnosis of heart abnormality detection.

RESULTS

As a result, the performance of DAE-ConvBiLSTM has obtained an average of above 98.59% accuracy, sensitivity, specificity, precision, and F1 score from the existing studies. The DAE-ConvBiLSTM has also experimented with detecting T-wave (due to ventricular repolarisation) morphology abnormalities.

CONCLUSION

The development architecture for detecting heart abnormalities using an unsupervised learning DAE and supervised learning ConvBiLSTM can be proposed for an end-to-end learning algorithm. In the future, the precise accuracy of the ECG main waveform will affect heart abnormalities detection in clinical practice.

Can the application of certain music information retrieval methods contribute to the machine learning classification of electrocardiographic signals?

The electrocardiogram is traditionally used to diagnose a large number of heart pathologies. Research to improve the readability and classification of cardiac signals includes studies geared toward sonification of the electrocardiographic signal and others involving features related to music processing, such as Mel-frequency cepstral coefficients. In terms of music processing features, this study seeks to use music information retrieval (MIR) features as electrocardiographic signal descriptors. The study compares the discriminatory capability of the introduced features in relation to standard groups such as heart rate variability, wavelet transform, descriptive statistics, Mel coefficients and fractal analysis, evaluated using classification algorithms; the signals analyzed were extracted from public databases. The group of features extracted from wavelet transform and the MIR group showed a high level of discrimination; the best representation of the ECG signals in the study was achieved in most cases by the MIR features. Moreover, a correlation coefficient higher than 0.8 was found between a number of MIR and other feature groups, indicating a likely relationship between the electrocardiographic signals and MIR features. These results suggest the feasibility of representing the analyzed signals by music information retrieval descriptors, giving the potential to consider these electrocardiographic signals as analogues to musical signals.

ECG Authentication Hardware Design With Low-Power Signal Processing and Neural Network Optimization With Low Precision and Structured Compression

Biometrics such as facial features, fingerprint, and iris are being used increasingly in modern authentication systems. These methods are now popular and have found their way into many portable electronics such as smartphones, tablets, and laptops. Furthermore, the use of biometrics enables secure access to private medical data, now collected in wearable devices such as smartwatches. In this work, we present an accurate low-power device authentication system that employs electrocardiogram (ECG) signals as the biometric modality. The proposed ECG processor consists of front-end signal processing of ECG signals and back-end neural networks (NNs) for accurate authentication. The NNs are trained using a cost function that minimizes intra-individual distance over time and maximizes inter-individual distance. Efficient low-power hardware was implemented by using fixed coefficients for ECG signal pre-processing and by using joint optimization of low-precision and structured sparsity for the NNs. We implemented two instances of ECG authentication hardware with 4X and 8X structurally-compressed NNs in 65 nm LP CMOS, which consume low power of 62.37  μW and 75.41  μW for real-time ECG authentication with a low equal error rate of 1.36% and 1.21%, respectively, for a large 741-subject in-house ECG database. The hardware was evaluated at 10 kHz clock frequency and 1.2 V voltage supply.

Low Resource Complexity R-peak Detection Based on Triangle Template Matching and Moving Average Filter

A novel R-peak detection algorithm suitable for wearable electrocardiogram (ECG) devices is proposed with four objectives: robustness to noise, low latency processing, low resource complexity, and automatic tuning of parameters. The approach is a two-pronged algorithm comprising (1) triangle template matching to accentuate the slope information of the R-peaks and (2) a single moving average filter to define a dynamic threshold for peak detection. The proposed algorithm was validated on eight ECG public databases. The obtained results not only presented good accuracy, but also low resource complexity, all of which show great potential for detection R-peaks in ECG signals collected from wearable devices.

Phase Relation between Depolarization and Repolarization Alternans in ECG

T-Wave Alternans (TWA) in the electro cardiogram (ECG) has been widely investigated as a potential predictor of ventricular arrhythmia. However, large clinical trials show that TWA has a high negative predictive value (NPV) but poor positive predictive value (PPV). Therefore, there is need for exploration of approaches to improve PPV of TWA. More recent studies suggest that whether alternans is spatially concordant or discordant affects arrhythmic potential. Results of our previous animal and simulation studies show that the phase relation between depolarization and repolarization alternans has an effect on the transition of concordant to discordant alternans. Towards the eventual goal of developing indexes that complement TWA and improve prediction of arrhythmia, the objectives in this study were to verify the existence of R wave amplitude alternans (RWAA, a surrogate of depolarization alternans) and investigate the phase relationship between RWAA and TWA in clinical grade ECGs. Results show that RWAA does occur in ECGs and that the phase relationship between RWAA and TWA can be labile. These results support further investigation of the co-occurrence of these alternans for prediction of arrhythmic events.

Heartbeats Do Not Make Good Pseudo-Random Number Generators: An Analysis of the Randomness of Inter-Pulse Intervals

The proliferation of wearable and implantable medical devices has given rise to an interest in developing security schemes suitable for these systems and the environment in which they operate. One area that has received much attention lately is the use of (human) biological signals as the basis for biometric authentication, identification and the generation of cryptographic keys. The heart signal (e.g., as recorded in an electrocardiogram) has been used by several researchers in the last few years. Specifically, the so-called Inter-Pulse Intervals (IPIs), which is the time between two consecutive heartbeats, have been repeatedly pointed out as a potentially good source of entropy and are at the core of various recent authentication protocols. In this work, we report the results of a large-scale statistical study to determine whether such an assumption is (or not) upheld. For this, we have analyzed 19 public datasets of heart signals from the Physionet repository, spanning electrocardiograms from 1353 subjects sampled at different frequencies and with lengths that vary between a few minutes and several hours. We believe this is the largest dataset on this topic analyzed in the literature. We have then applied a standard battery of randomness tests to the extracted IPIs. Under the algorithms described in this paper and after analyzing these 19 public ECG datasets, our results raise doubts about the use of IPI values as a good source of randomness for cryptographic purposes. This has repercussions both in the security of some of the protocols proposed up to now and also in the design of future IPI-based schemes.

CSE database: extended annotations and new recommendations for ECG software testing

Nowadays, cardiovascular diseases represent the most common cause of death in western countries. Among various examination techniques, electrocardiography (ECG) is still a highly valuable tool used for the diagnosis of many cardiovascular disorders. In order to diagnose a person based on ECG, cardiologists can use automatic diagnostic algorithms. Research in this area is still necessary. In order to compare various algorithms correctly, it is necessary to test them on standard annotated databases, such as the Common Standards for Quantitative Electrocardiography (CSE) database. According to Scopus, the CSE database is the second most cited standard database. There were two main objectives in this work. First, new diagnoses were added to the CSE database, which extended its original annotations. Second, new recommendations for diagnostic software quality estimation were established. The ECG recordings were diagnosed by five new cardiologists independently, and in total, 59 different diagnoses were found. Such a large number of diagnoses is unique, even in terms of standard databases. Based on the cardiologists' diagnoses, a four-round consensus (4R consensus) was established. Such a 4R consensus means a correct final diagnosis, which should ideally be the output of any tested classification software. The accuracy of the cardiologists' diagnoses compared with the 4R consensus was the basis for the establishment of accuracy recommendations. The accuracy was determined in terms of sensitivity = 79.20-86.81%, positive predictive value = 79.10-87.11%, and the Jaccard coefficient = 72.21-81.14%, respectively. Within these ranges, the accuracy of the software is comparable with the accuracy of cardiologists. The accuracy quantification of the correct classification is unique. Diagnostic software developers can objectively evaluate the success of their algorithm and promote its further development. The annotations and recommendations proposed in this work will allow for faster development and testing of classification software. As a result, this might facilitate cardiologists' work and lead to faster diagnoses and earlier treatment.

TERMA Framework for Biomedical Signal Analysis: An Economic-Inspired Approach

Biomedical signals contain features that represent physiological events, and each of these events has peaks. The analysis of biomedical signals for monitoring or diagnosing diseases requires the detection of these peaks, making event detection a crucial step in biomedical signal processing. Many researchers have difficulty detecting these peaks to investigate, interpret and analyze their corresponding events. To date, there is no generic framework that captures these events in a robust, efficient and consistent manner. A new method referred to for the first time as two event-related moving averages ("TERMA") involves event-related moving averages and detects events in biomedical signals. The TERMA framework is flexible and universal and consists of six independent LEGO building bricks to achieve high accuracy detection of biomedical events. Results recommend that the window sizes for the two moving averages ( W 1 and W 2 ) have to follow the inequality ( 8 × W 1 ) ≥ W 2 ≥ ( 2 × W 1 ) . Moreover, TERMA is a simple yet efficient event detector that is suitable for wearable devices, point-of-care devices, fitness trackers and smart watches, compared to more complex machine learning solutions.

Tensor-based detection of T wave alternans using ECG

T wave alternans is defined as changes in the T wave amplitude in an ABABAB-pattern. It can be found in ECG signals of patients with heart diseases and is a possible indicator to predict the risk on sudden cardiac death. Due to its low amplitude, robust automatic T wave alternans detection is a difficult task. We present a new method to detect T wave alternans in multichannel ECG signals. The use of tensors (multidimensional matrices) permits the combination of the information present in different channels, making detection more reliable. The possibility of decomposition of incomplete tensors is exploited to deal with noisy ECG segments. Using a sliding window of 128 heartbeats, a tensor is constructed of the T waves of all channels. Canonical Polyadic Decomposition is applied to this tensor and the resulting loading vectors are examined for information about the T wave behavior in three dimensions. T wave alternans is detected using a sign change counting method that is able to extract both the T wave alternans length and magnitude. When applying this novel method to a database of patients with multiple positive T wave alternans tests using the clinically available spectral method tests, both the length and the magnitude of the detected T wave alternans is larger for these subjects than for subjects in a control group.

A time-domain hybrid analysis method for detecting and quantifying T-wave alternans

T-wave alternans (TWA) in surface electrocardiograph (ECG) signals has been recognized as a marker of cardiac electrical instability and is hypothesized to be associated with increased risk for ventricular arrhythmias among patients. A novel time-domain TWA hybrid analysis method (HAM) utilizing the correlation method and least squares regression technique is described in this paper. Simulated ECGs containing artificial TWA (cases of absence of TWA and presence of stationary or time-varying or phase-reversal TWA) under different baseline wanderings are used to test the method, and the results show that HAM has a better ability of quantifying TWA amplitude compared with the correlation method (CM) and adapting match filter method (AMFM). The HAM is subsequently used to analyze the clinical ECGs, and results produced by the HAM have, in general, demonstrated consistency with those produced by the CM and the AMFM, while the quantifying TWA amplitudes by the HAM are universally higher than those by the other two methods.

Fast QRS detection with an optimized knowledge-based method: evaluation on 11 standard ECG databases

The current state-of-the-art in automatic QRS detection methods show high robustness and almost negligible error rates. In return, the methods are usually based on machine-learning approaches that require sufficient computational resources. However, simple-fast methods can also achieve high detection rates. There is a need to develop numerically efficient algorithms to accommodate the new trend towards battery-driven ECG devices and to analyze long-term recorded signals in a time-efficient manner. A typical QRS detection method has been reduced to a basic approach consisting of two moving averages that are calibrated by a knowledge base using only two parameters. In contrast to high-accuracy methods, the proposed method can be easily implemented in a digital filter design.

A robust and simple security extension for the medical standard SCP-ECG

This paper proposes a SCP-ECG security extension after having analyzed the features of this standard, its security requirements and the current measures implemented by other medical protocols. Our approach permits SCP-ECG files to be stored safely and proper access to be granted (or denied) to users for different purposes: interpretation of the test, consultation, clinical research or teaching. The access privileges are scaled by means of role-based profiles supported by cryptographic elements (ciphering, digital certificates and digital signatures). These elements are arranged as metadata into a new section which extends the protocol and protects the remaining sections. The application built to implement this approach has been extensively tested, showing its capacity to authenticate users and to protect the integrity of files and the privacy of sensitive data, with a low impact on file size and access time. In addition, this solution is compatible with any version of the SCP-ECG and can be easily integrated into e-health platforms.

Comparison of three T-Wave Delineation Algorithms based on Wavelet Filterbank, Correlation and PCA

There is a large interest in analysing the QT-interval, as a prolonged QT-interval can cause the development of ventricular tachyarrhythmias such as Torsade de Pointes. One major part of QT-analysis is T-end detection. Three automatic T-end delineation methods based on wavelet filterbanks (WAM), correlation (CORM) and Principal Component Analysis PCA (PCAM) have been developed and applied to Physionet QT database.All algorithms tested on Physionet QT database showed good results, while PCAM produced better results than WAM and CORM achieved best results. Standard deviation in sampling points (f(s)=250Hz) have been 33.3 (WAM), 8.0 (PTDM) and 7.8 (CORM). It could be shown that WAM is prone to interference while CORM is the most stable method even under bad conditions. Furthermore it was possible to detect significant QT-prolongation caused by Moxifloxacin in Thorough QT Study # 2 using CORM. QT-prolongation is significantly correlated to blood plasma concentration of Moxifloxacin.

An artificial vector model for generating abnormal electrocardiographic rhythms

We present generalizations of our previously published artificial models for generating multi-channel ECG to provide simulations of abnormal cardiac rhythms. Using a three-dimensional vectorcardiogram (VCG) formulation, we generate the normal cardiac dipole for a patient using a sum of Gaussian kernels, fitted to real VCG recordings. Abnormal beats are specified either as perturbations to the normal dipole or as new dipole trajectories. Switching between normal and abnormal beat types is achieved using a first-order Markov chain. Probability transitions can be learned from real data or modeled by coupling to heart rate and sympathovagal balance. Natural morphology changes from beat-to-beat are incorporated by varying the angular frequency of the dipole as a function of the inter-beat (RR) interval. The RR interval time series is generated using our previously described model whereby time- and frequency-domain heart rate (HR) and heart rate variability characteristics can be specified. QT-HR hysteresis is simulated by coupling the Gaussian kernels associated with the T-wave in the model with a nonlinear factor related to the local HR (determined from the last n RR intervals). Morphology changes due to respiration are simulated by introducing a rotation matrix couple to the respiratory frequency. We demonstrate an example of the use of this model by simulating HR-dependent T-wave alternans (TWA) with and without phase-switching due to ectopy. Application of our model also reveals previously unreported effects of common TWA estimation methods.

An Open-Source Standard T-Wave Alternans Detector for Benchmarking

We describe an open source algorithm suite for T-Wave Alternans (TWA) detection and quantification. The software consists of Matlab implementations of the widely used Spectral Method and Modified Moving Average with libraries to read both WFDB and ASCII data under windows and Linux. The software suite can run in both batch mode and with a provided graphical user interface to aid waveform exploration. Our software suite was calibrated using an open source TWA model, described in a partner paper [1] by Clifford and Sameni. For the PhysioNet/CinC Challenge 2008 we obtained a score of 0.881 for the Spectral Method and 0.400 for the MMA method. However, our objective was not to provide the best TWA detector, but rather a basis for detailed discussion of algorithms.

An Artificial Multi-Channel Model for Generating Abnormal Electrocardiographic Rhythms

We present generalizations of our previously published artificial models for generating multi-channel ECG so that the simulation of abnormal rhythms is possible. Using a three-dimensional vectorcardiogram (VCG) formulation, we generate the normal cardiac dipole for a patient using a sum of Gaussian kernels, fitted to real VCG recordings. Abnormal beats are then specified either as new dipoles, or as perturbations of the existing dipole. Switching between normal and abnormal beat types is achieved using a hidden Markov model (HMM). Probability transitions can be learned from real data or modeled by coupling to heart rate and sympathovagal balance. Natural morphology changes form beat-to-beat are incorporated as before from varying the angular frequency of the dipole as a function of the inter-beat (RR) interval. The RR interval time series is generated using our previously described model whereby time-and frequency-domain heart rate (HR) and heart rate variability (HRV) characteristics can be specified. QT-HR hysteresis is simulated by coupling the Gaussian kernels associated with the T-wave in the model with a nonlinear factor related to the local HR (determined from the last n RR intervals). Morphology changes due to respiration are simulated by coupling the RR interval to the angular frequency of the dipole. We demonstrate an example of the use of this model by simulating T-Wave Alternans (TWA). The magnitude of the TWA effect is modeled as a disturbance on the T-loop of the dipole with a magnitude that differs in each of the three VCG planes. The effect is then turned on or off using a HMM. The values of the transition matrix are determined by the local heart rate, such that when the HR ramps up towards 100 BPM, the probability of observing a TWA effect rapidly but smoothly increases. In this way, no 'sudden' switching from non-TWA to TWA is observed, and the natural tendency for TWA to be associated with a critical HR-related activation level is simulated. Finally, to generate multi-lead signals, the VCG is mapped to any set of clinical leads using a Dower-like transform derived from a least-squares optimization between known VCGs and known lead morphologies. ECGs with calibrated amounts of TWA were generated by this model and included in the PhysioNet/CinC Challenge 2008 data set.