Performance Analysis of Ten Common QRS Detectors on Different ECG Application Cases

Quantifying the Suitability of Biosignals Acquired During Surgery for Multimodal Analysis

Goal: Recently, large datasets of biosignals acquired during surgery became available. As they offer multiple physiological signals measured in parallel, multimodal analysis - which involves their joint analysis - can be conducted and could provide deeper insights than unimodal analysis based on a single signal. However, it is unclear what percentage of intraoperatively acquired data is suitable for multimodal analysis. Due to the large amount of data, manual inspection and labelling into suitable and unsuitable segments are not feasible. Nevertheless, multimodal analysis is performed successfully in sleep studies since many years as their signals have proven suitable. Hence, this study evaluates the suitability to perform multimodal analysis on a surgery dataset (VitalDB) using a multi-center sleep dataset (SIESTA) as reference. Methods: We applied widely known algorithms entitled "signal quality indicators" to the common biosignals in both datasets, namely electrocardiography, electroencephalography, and respiratory signals split in segments of 10 s duration. As there are no multimodal methods available, we used only unimodal signal quality indicators. In case, all three signals were determined as being adequate by the indicators, we assumed that the whole signal segment was suitable for multimodal analysis. Results: 82% of SIESTA and 72% of VitalDB are suitable for multimodal analysis. Unsuitable signal segments exhibit constant or physiologically unreasonable values. Histogram examination indicated similar signal quality distributions between the datasets, albeit with potential statistical biases due to different measurement setups. Conclusions: The majority of data within VitalDB is suitable for multimodal analysis.

QRS Detector Performance Evaluation Aware of Temporal Accuracy and Presence of Noise

Algorithms for QRS detection are fundamental in the ECG interpretive processing chain. They must meet several challenges, such as high reliability, high temporal accuracy, high immunity to noise, and low computational complexity. Unfortunately, the accuracy expressed by missed or redundant events statistics is often the only parameter used to evaluate the detector's performance. In this paper, we first notice that statistics of true positive detections rely on researchers' arbitrary selection of time tolerance between QRS detector output and the database reference. Next, we propose a multidimensional algorithm evaluation method and present its use on four example QRS detectors. The dimensions are (a) influence of detection temporal tolerance, tested for values between 8.33 and 164 ms; (b) noise immunity, tested with an ECG signal with an added muscular noise pattern and signal-to-noise ratio to the effect of "no added noise", 15, 7, 3 dB; and (c) influence of QRS morphology, tested on the six most frequently represented morphology types in the MIT-BIH Arrhythmia Database. The multidimensional evaluation, as proposed in this paper, allows an in-depth comparison of QRS detection algorithms removing the limitations of existing one-dimensional methods. The method enables the assessment of the QRS detection algorithms according to the medical device application area and corresponding requirements of temporal accuracy, immunity to noise, and QRS morphology types. The analysis shows also that, for some algorithms, adding muscular noise to the ECG signal improves algorithm accuracy results.

Stress influence on real-world driving identified by monitoring heart rate variability and morphologic variability of electrocardiogram signals: the case of intercity roads

Objectives. This study examines which of the heart rate variability (HRV) and morphologic variability (MV) metrics may have the highest accuracy in different stress detection during real-world driving. Methods. The cross-sectional study was carried out among 93 intercity mini-bus male drivers aged 22-67 years. The Trillium 5000 Holter Recorder and GARMIN Virb Elite camera were used to determine heart rate and vehicle speed measurements along the path, respectively. We considered the HRV and MV metrics of electrocardiogram (ECG) signals including the mean RR interval (mRR), mean heart rate (mHR), normalized low-frequency spectrum (nLF), normalized high-frequency spectrum (nHF), normalized very low-frequency spectrum (nVLF), difference of normalized low-frequency spectrum and normalized high-frequency spectrum (dLFHF), and sympathovagal balance index (SVI). Results. The analysis showed that the HRV metrics mHR, mRR, nVLF, nLF, nHF, dLFHF and SVI are effective in mental stress detection while driving as compared to rest time. We obtained a high accuracy of stress detection for MV metrics as compared to the traditional HRV analysis, of approximately 92%. Conclusions. Our findings indicate that driver stress could be detected with an accuracy of 92% using MV metrics as an accurate physiological index of the driver's state.

Heartbeat detection from high-density EMG electrodes on the upper arm at different EMG intensity levels using Zephlet

BACKGROUND AND OBJECTIVES

A significant number of global deaths caused by cardiac arrhythmias can be prevented with accurate and immediate identification. Wearable devices can play a critical role in such identification by continuously monitoring cardiac activity using electrocardiogram (ECG). The existing body of research has focused on extracting cardiac information from the body surface by investigating various electrode locations and algorithm development for ECG interpretation. The present study was designed for heartbeat detection using the signals recorded from the upper arm.

METHODS

Firstly, optimal electrode locations on the upper arm were identified for Rest and elbow flexion (EF) conditions. Next, a synthesized ECG was generated using the selected electrodes with generalized weights over subjects and trials, and then zero-phase wavelet (Zephlet) was applied for feature extraction. Heartbeat detection was finally performed using the extracted detail coefficients incorporated with a multiagent detection scheme (MDS).

RESULTS

The F1-score for heartbeat detection was 0.94  ±  0.16, 0.86  ±  0.22, 0.79  ±  0.26, and 0.67  ±  0.31 for Rest and EF with three different levels of muscle contraction (C1 to C3), respectively. Changing the acceptable distance between the detected and actual heartbeats from 50 ms to 20 ms, the F1-score changed to 0.81  ±  0.20, 0.66  ±  0.26, 0.57  ±  0.26,  and 0.44  ±  0.26 for Rest and C1 to C3, respectively.

CONCLUSION

These findings make several contributions to the current literature, summarized as precise and consistent electrode localization for various muscle contraction levels and accurate heartbeat detection method development for each of these conditions.

Robust Peak Detection for Holter ECGs by Self-Organized Operational Neural Networks

Although numerous R-peak detectors have been proposed in the literature, their robustness and performance levels may significantly deteriorate in low-quality and noisy signals acquired from mobile electrocardiogram (ECG) sensors, such as Holter monitors. Recently, this issue has been addressed by deep 1-D convolutional neural networks (CNNs) that have achieved state-of-the-art performance levels in Holter monitors; however, they pose a high complexity level that requires special parallelized hardware setup for real-time processing. On the other hand, their performance deteriorates when a compact network configuration is used instead. This is an expected outcome as recent studies have demonstrated that the learning performance of CNNs is limited due to their strictly homogenous configuration with the sole linear neuron model. This has been addressed by operational neural networks (ONNs) with their heterogenous network configuration encapsulating neurons with various nonlinear operators. In this study, to further boost the peak detection performance along with an elegant computational efficiency, we propose 1-D Self-Organized ONNs (Self-ONNs) with generative neurons. The most crucial advantage of 1-D Self-ONNs over the ONNs is their self-organization capability that voids the need to search for the best operator set per neuron since each generative neuron has the ability to create the optimal operator during training. The experimental results over the China Physiological Signal Challenge-2020 (CPSC) dataset with more than one million ECG beats show that the proposed 1-D Self-ONNs can significantly surpass the state-of-the-art deep CNN with less computational complexity. Results demonstrate that the proposed solution achieves a 99.10% F1-score, 99.79% sensitivity, and 98.42% positive predictivity in the CPSC dataset, which is the best R-peak detection performance ever achieved.

A Generative Adversarial Network to Synthesize 3D Magnetohydrodynamic Distortions for Electrocardiogram Analyses Applied to Cardiac Magnetic Resonance Imaging

Recently, deep learning (DL) models have been increasingly adopted for automatic analyses of medical data, including electrocardiograms (ECGs). Large, available ECG datasets, generally of high quality, often lack specific distortions, which could be helpful for enhancing DL-based algorithms. Synthetic ECG datasets could overcome this limitation. A generative adversarial network (GAN) was used to synthesize realistic 3D magnetohydrodynamic (MHD) distortion templates, as observed during magnetic resonance imaging (MRI), and then added to available ECG recordings to produce an augmented dataset. Similarity metrics, as well as the accuracy of a DL-based R-peak detector trained with and without data augmentation, were used to evaluate the effectiveness of the synthesized data. Three-dimensional MHD distortions produced by the proposed GAN were similar to the measured ones used as input. The precision of a DL-based R-peak detector, tested on actual unseen data, was significantly enhanced by data augmentation; its recall was higher when trained with augmented data. Using synthesized MHD-distorted ECGs significantly improves the accuracy of a DL-based R-peak detector, with a good generalization capacity. This provides a simple and effective alternative to collecting new patient data. DL-based algorithms for ECG analyses can suffer from bias or gaps in training datasets. Using a GAN to synthesize new data, as well as metrics to evaluate its performance, can overcome the scarcity issue of data availability.

Precise detection and localization of R-peaks from ECG signals

Heart rate variability (HRV) is derived from the R-R interval, which depends on the precise localization of R-peaks within an electrocardiogram (ECG) signal. However, current algorithm assessment methods prioritize the R-peak detection's sensitivity rather than the precision of pinpointing the exact R-peak positions. As a result, it is of great value to develop an R-peak detection algorithm with high-precision R-peak localization. This paper introduces a novel R-peak localization algorithm that involves modifications to the well-established Pan-Tompkins (PT) algorithm. The algorithm was implemented as follows. First, the raw ECG signal $ X\left(i\right) $ was band-pass filtered (5-35 Hz) to obtain a preprocessed signal $ Y\left(i\right) $. Second, $ Y\left(i\right) $ was squared to enhance the QRS complex, followed by a 5 Hz low-pass filter to obtain the QRS envelope, which was transformed into a window signal $ W\left(i\right) $ by dynamic threshold with a minimum width of 200 ms to mark the QRS complex. Third, $ Y\left(i\right) $ was used to generate QRS template $ T\left(n\right) $ automatically, and then the R-peak was identified by a template matching process to find the maximum absolute value of all cross-correlation values between $ T\left(n\right) $ and $ Y\left(i\right) $. The proposed algorithm achieved a sensitivity (SE) of 99.78%, a positive prediction value (PPV) of 99.78% and data error rate (DER) of 0.44% in R-peak localization for the MIT-BIH Arrhythmia database. The annotated-detected error (ADE), which represents the error between the annotated R-peak location and the detected R-peak location, was 8.35 ms for the MIT-BIH Arrhythmia database. These results outperformed the results obtained using the classical Pan-Tompkins algorithm which yielded an SE of 98.87%, a PPV of 99.14%, a DER of 1.98% and an ADE of 21.65 ms for the MIT-BIH Arrhythmia database. It can be concluded that the algorithm can precisely detect the location of R-peaks and may have the potential to enhance clinical applications of HRV analysis.

Completing the Cabrera Circle: deriving adaptable leads from ECG limb leads by combining constraints with a correction factor

Objective.We present a concept for processing 6-lead electrocardiography (ECG) signals which can be applied to various use cases in quantitative electrocardiography.Approach.Our work builds upon the mathematics of the well-known Cabrera sequence which is a re-sorting of the six limb leads (I,II,III,aVR,aVL,aVF) into a clockwise and physiologically-interpretable order. By deriving correction factors for harmonizing lead strengths and choosing an appropriate basis for the leads, we extend this concept towards what we call the 'Cabrera Circle' based on a mathematically sound foundation.Main results.To demonstrate the practical effectiveness and relevance of this concept, we analyze its suitability for deriving interpolated leads between the six limb leads and a 'radial' lead which both can be useful for specific use cases. We focus on the use cases of i) determination of the electrical heart axis by proposing a novel interactive tool for reconstructing the heart's vector loop and ii) improving accuracy in time of automatic R-wave detection and T-wave delineation in 6-lead ECG. For the first use case, we derive an equation which allows projections of the 2-dimensional vector loops to arbitrary angles of the Cabrera Circle. For the second use case, we apply several state-of-the-art algorithms to a freely-available 12-lead dataset (Lobachevsky University Database). Out-of-the-box results show that the derived radial lead outperforms the other limb leads (I,II,III,aVR,aVL,aVF) by improving F1 scores of R-peak and T-peak detection by 0.61 and 2.12, respectively. Results of on- and offset computations are also improved but on a smaller scale.Significance.In summary, the Cabrera Circle offers a methodology that might be useful for quantitative electrocardiography of the 6-lead subsystem-especially in the digital age.

The truth Hertz-synchronization of electroencephalogram signals with physiological waveforms recorded in an intensive care unit

Objective.The ability to synchronize continuous electroencephalogram (cEEG) signals with physiological waveforms such as electrocardiogram (ECG), invasive pressures, photoplethysmography and other signals can provide meaningful insights regarding coupling between brain activity and other physiological subsystems. Aligning these datasets is a particularly challenging problem because device clocks handle time differently and synchronization protocols may be undocumented or proprietary.Approach.We used an ensemble-based model to detect the timestamps of heartbeat artefacts from ECG waveforms recorded from inpatient bedside monitors and from cEEG signals acquired using a different device. Vectors of inter-beat intervals were matched between both datasets and robust linear regression was applied to measure the relative time offset between the two datasets as a function of time.Main Results.The timing error between the two unsynchronized datasets ranged between -84 s and +33 s (mean 0.77 s, median 4.31 s, IQR25-4.79 s, IQR75 11.38s). Application of our method improved the relative alignment to within ± 5ms for more than 61% of the dataset. The mean clock drift between the two datasets was 418.3 parts per million (ppm) (median 414.6 ppm, IQR25 411.0 ppm, IQR75 425.6 ppm). A signal quality index was generated that described the quality of alignment for each cEEG study as a function of time.Significance.We developed and tested a method to retrospectively time-align two clinical waveform datasets acquired from different devices using a common signal. The method was applied to 33,911h of signals collected in a paediatric critical care unit over six years, demonstrating that the method can be applied to long-term recordings collected under clinical conditions. The method can account for unknown clock drift rates and the presence of discontinuities caused by clock resynchronization events.

A Comprehensive Comparison of Six Publicly Available Algorithms for Localization of QRS Complex on Electrocardiograph

The QRS complex is the most prominent feature of the electrocardiogram (ECG) that is used as a marker to identify the cardiac cycles. Identification of QRS complex locations enables arrhythmia detection and heart rate variability estimation. Therefore, accurate and consistent localization of the QRS complex is an important component of automated ECG analysis which is necessary for the early detection of cardiovascular diseases. This study evaluates the performance of six popular publicly available QRS complex detection methods on a large dataset of over half a million ECGs in a diverse population of patients. We found that a deep-learning method that won first place in the 2019 Chinese physiological challenge (CPSC-1) outperforms the remaining five QRS complex detection methods with an F1 score of 98.8% and an absolute sdRR error of 5.5 ms. We also examined the stratified performance of the studied methods on various cardiac conditions. All six methods had a lower performance in the detection of QRS complexes in ECG signals of patients with pacemakers, complete atrioventricular block, or indeterminate cardiac axis. We also concluded that, in the presence of different cardiac conditions, CPSC-1 is more robust than Pan-Tompkins which is the most popular model for QRS complex detection. We expect that this study can potentially serve as a guide for researchers on the appropriate QRS detection method for their target applications.Clinical Relevance-This study highlights the overall performance of publicly available QRS detection algorithms in a large dataset of diverse patients. We showed that there are specific cardiac conditions that are associated with the poor performance of QRS detection algorithms and may adversely influence the performance of algorithms that rely on accurate and reliable QRS detection.

A Deep Learning Architecture Using 3D Vectorcardiogram to Detect R-Peaks in ECG with Enhanced Precision

Providing reliable detection of QRS complexes is key in automated analyses of electrocardiograms (ECG). Accurate and timely R-peak detections provide a basis for ECG-based diagnoses and to synchronize radiologic, electrophysiologic, or other medical devices. Compared with classical algorithms, deep learning (DL) architectures have demonstrated superior accuracy and high generalization capacity. Furthermore, they can be embedded on edge devices for real-time inference. 3D vectorcardiograms (VCG) provide a unifying framework for detecting R-peaks regardless of the acquisition strategy or number of ECG leads. In this article, a DL architecture was demonstrated to provide enhanced precision when trained and applied on 3D VCG, with no pre-processing nor post-processing steps. Experiments were conducted on four different public databases. Using the proposed approach, high F1-scores of 99.80% and 99.64% were achieved in leave-one-out cross-validation and cross-database validation protocols, respectively. False detections, measured by a precision of 99.88% or more, were significantly reduced compared with recent state-of-the-art methods tested on the same databases, without penalty in the number of missed peaks, measured by a recall of 99.39% or more. This approach can provide new applications for devices where precision, or positive predictive value, is essential, for instance cardiac magnetic resonance imaging.

Algorithm for Mobile Platform-Based Real-Time QRS Detection

Recent advancements in smart, wearable technologies have allowed the detection of various medical conditions. In particular, continuous collection and real-time analysis of electrocardiogram data have enabled the early identification of pathologic cardiac rhythms. Various algorithms to assess cardiac rhythms have been developed, but these utilize excessive computational power. Therefore, adoption to mobile platforms requires more computationally efficient algorithms that do not sacrifice correctness. This study presents a modified QRS detection algorithm, the AccYouRate Modified Pan-Tompkins (AMPT), which is a simplified version of the well-established Pan-Tompkins algorithm. Using archived ECG data from a variety of publicly available datasets, relative to the Pan-Tompkins, the AMPT algorithm demonstrated improved computational efficiency by 5-20×, while also universally enhancing correctness, both of which favor translation to a mobile platform for continuous, real-time QRS detection.

Internet of Things-Based ECG and Vitals Healthcare Monitoring System

Health monitoring and its associated technologies have gained enormous importance over the past few years. The electrocardiogram (ECG) has long been a popular tool for assessing and diagnosing cardiovascular diseases (CVDs). Since the literature on ECG monitoring devices is growing at an exponential rate, it is becoming difficult for researchers and healthcare professionals to select, compare, and assess the systems that meet their demands while also meeting the monitoring standards. This emphasizes the necessity for a reliable reference to guide the design, categorization, and analysis of ECG monitoring systems, which will benefit both academics and practitioners. We present a complete ECG monitoring system in this work, describing the design stages and implementation of an end-to-end solution for capturing and displaying the patient's heart signals, heart rate, blood oxygen levels, and body temperature. The data will be presented on an OLED display, a developed Android application as well as in MATLAB via serial communication. The Internet of Things (IoT) approaches have a clear advantage in tackling the problem of heart disease patient care as they can transform the service mode into a widespread one and alert the healthcare services based on the patient's physical condition. Keeping this in mind, there is also the addition of a web server for monitoring the patient's status via WiFi. The prototype, which is compliant with the electrical safety regulations and medical equipment design, was further benchmarked against a commercially available off-the-shelf device, and showed an excellent accuracy of 99.56%.

Robust R-peak detection in an electrocardiogram with stationary wavelet transformation and separable convolution

R-peak detection is an essential step in analyzing electrocardiograms (ECGs). Previous deep learning models reported their performance primarily in a single database, and some models did not perform at the highest levels when applied to a database different from the testing database. To achieve high performances in cross-database validations, we developed a novel deep learning model for R-peak detection using stationary wavelet transform (SWT) and separable convolution. Three databases (i.e., the MIT-BIH Arrhythmia [MIT-BIH], the Institute of Cardiological Technics [INCART], and the QT) were used in both the training and testing models, and the MIT-BIH ST Change (MIT-BIH-ST), European ST-T, TELE and MIT-BIH Noise Stress Test (MIT-BIH-NST) databases were further used for testing. The detail coefficient of level 4 decomposition by SWT and the first derivative from filtered ECGs were used for model inputs, and the interval of 150 ms centered at marked peaks was used for labels. Separable convolution with atrous spatial pyramidal pooling was selected as the model's architecture, and noise-augmented waveforms of 5.69 s duration (2048 size in 360 Hz) were used in training. The model performance was evaluated using cross-database validation. The F1 scores of the peak detection model were 0.9994, 0.9985, and 0.9999 in the MIT-BIH, INCART, and QT databases, respectively. When the above three databases were pooled, the F1 scores were 0.9993 for fivefold cross-validation and 0.9991 for cross-database validation. The model performance remained high for MIT-BIH-ST, European ST-T, and TELE, with F1 scores of 0.9995, 0.9988, and 0.9790, respectively. The model performance when trained by severe noise augmentation increased for the MIT-BIH-NST database (F1 scores from 0.9504 to 0.9759) and decreased for the MIT-BIH database (F1 scores from 0.9994 to 0.9991). The present SWT and separable convolution-based model for R-peak detection yields a high performance even for cross-database validations.

A comparative study on neural networks for paroxysmal atrial fibrillation events detection from electrocardiography

OBJECTIVE

This work conducts a comparative study on the effect of neural networks of different architectures on the detection of paroxysmal atrial fibrillation (PAF) events from dynamic electrocardiography (ECG) recordings, a problem raised in the 4th China Physiological Signal Challenge 2021 (CPSC2021).

APPROACH

We proposed 3 neural network models and an auxiliary one for QRS detection to tackle the problem. A convolutional recurrent neural network (CRNN) model and a U-Net model that accepts ECG waveform input make sample-wise predictions. This regards the PAF events detection as a segmentation task. A stacked bidirectional long short-term memory (LSTM) model takes the sequence of RR intervals, which is derived from the output of the QRS detection model and makes beat-wise predictions. The QRS detection model also has a CRNN architecture, which is slightly different from the model for the AF segmentation task. Final predictions are merged by outputs from models making sample-wise predictions and making beat-wise predictions. Finally, the locations of QRS complexes are used to filter out segments (both normal and AF) shorter than 5 beats. In order to make the neural network models more sensitive to the critical sample points (onsets and offsets) of the AF events, we proposed a novel masked binary cross-entropy (MaskedBCE) loss function for training the models. This loss function is the conventional BCE loss multiplied by a mask, whose values in a neighbourhood of critical sample points are significantly larger than elsewhere.

MAIN RESULTS

Our method received a score of 1.9972 on the first part of the hidden test set of CPSC2021 and a score of 3.0907 on the second part. The average score was 2.5440, ranked 5th out of 17 teams with successful official entries.

SIGNIFICANCE

This work proposed an effective solution to the problem of the detection of PAF events from dynamic ECGs and validated the efficacy of several neural network architectures on this problem.

A two-step pre-processing tool to remove Gaussian and ectopic noise for heart rate variability analysis

Artifacts in the Electrocardiogram (ECG) degrade the quality of the recorded signal and are not conducive to heart rate variability (HRV) analysis. The two types of noise most often found in ECG recordings are technical and physiological artifacts. Current preprocessing methods primarily attend to ectopic beats but do not consider technical issues that affect the ECG. A secondary aim of this study was to investigate the effect of increasing increments of artifacts on 24 of the most used HRV measures. A two-step preprocessing approach for denoising HRV is introduced which targets each type of noise separately. First, the technical artifacts in the ECG are eliminated by applying complete ensemble empirical mode decomposition with adaptive noise. The second step removes physiological artifacts from the HRV signal using a combination filter of single dependent rank order mean and an adaptive filtering algorithm. The performance of the two-step pre-processing tool showed a high correlation coefficient of 0.846 and RMSE value of 7.69 × 10^-5 for 6% of added ectopic beats and 6 dB Gaussian noise. All HRV measures studied except HF peak and LF peak are significantly affected by both types of noise. Frequency measures of Total power, HF power, and LF power and fragmentation measures; PAS, PIP, and PSS are the most sensitive to both types of noise.

Heartbeat detection from the upper arm using an SWT-based zero-phase filter bank incorporated with a voting Scheme

Electrocardiogram (ECG) signal provides a graphical representation of cardiac activity and is the most commonly adopted clinical tool for cardiac abnormalities detection. Heartbeat detection, as the first step in analyzing ECG signals, is required for an accurate diagnosis. Stationary wavelet transform (SWT) as a commonly used algorithm for heartbeat detection has a disadvantage of phase shift regarding the original signal. This work addresses this issue by presenting a new method that incorporates an SWT-based zero-phase filter bank with a voting scheme. Our results indicated that a superior performance in heartbeat detection was achieved from the upper arm compared to conventional SWT with a more accurate localization. We achieved sensitivity (SE) and positive predictive value (PPV) of 0.98±0.04 and 0.95±0.09 with the most distance of 50 ms from the actual heartbeats. The SE and PPV changed to 0.75±0.15 and 0.73±0.16, respectively for the distance of 20 ms. Clinical Relevance- The proposed method can be later implemented in wearable devices for convenient cardiac activity monitoring from upper arm or other none-conventional sites.

Lightweight heartbeat detection algorithm for consumer grade wearable ECG measurement devices and its implementation

We have developed a lightweight heartbeat detection algorithm for electrocardiogram (ECG) measurement in consumer-grade wearable devices. Based on a principle of calculating the clearance height in the inverted time differential of an ECG, the algorithm has a sensitivity specific to the QRS complex in an ECG. Because this algorithm is simple and lightweight, an efficient implementation form has been established for porting the algorithm to consumer-grade devices. We experimentally confirmed that the algorithm runs in an actual commercialized wearable device and provides improvement of heart rate measurement with noise caused by body motion.

Supraventricular ectopic beats and ventricular ectopic beats detection based on improved U-net

OBJECTIVE

Supraventricular ectopic beats (SVEB) or ventricular ectopic beats (VEB) are common arrhythmia with uncertain occurrence and morphological diversity, so realizing their automatic localization is of great significance in clinical diagnosis.

METHODS

We propose a modified U-net network: USV-net, it can simultaneously realize the automatic positioning of VEB and SVEB. The improvement consists of three parts: Firstly, we reconstruct part of the convolutional layer in U-net using group convolution to reduce the expression of redundant features. Secondly, a plug-and-play multi-scale 2D deformable convolution (MSDC) module is designed to extract cross-channel features of different scales. Thirdly, in addition to conventional output of U-net, we also compress and output the bottom feature map of U-net, the dual-output is trained through Dice-loss to take into account the learning of high/low resolution features of the model. We used the MIT-BIH arrhythmia database for patient-specific training, and used Sensitivity, Positive prediction rate and F1-scores to evaluate the effectiveness of our method.

MAIN RESULT

The F1-scores of SVEB and VEB achieve the best results compared with other studies in different testing records. It is worth noting that the F1-scores of SVEB and VEB reached 81.3 and 95.4 in the 24 testing records. Moreover, our method is also at the forefront in Sensitivity and Positive prediction rate.

SIGNIFICANCE

The method proposed in this paper has great potential in the detection of SVEB and VEB. We anticipate efficiency and accuracy of clinical detection of ectopic beats would be improved.

A Heart Rate Variability-Based Paroxysmal Atrial Fibrillation Prediction System

Atrial fibrillation (AF) is characterized by totally disorganized atrial depolarizations without effective atrial contraction. It is the most common form of cardiac arrhythmia, affecting more than 46.3 million people worldwide and its incidence rate remains increasing. Although AF itself is not life-threatening, its complications, such as strokes and heart failure, are lethal. About 25% of paroxysmal AF (PAF) patients become chronic for an observation period of more than one year. For long-term and real-time monitoring, a PAF prediction system was developed with four objectives: (1) high prediction accuracy, (2) fast computation, (3) small data storage, and (4) easy medical interpretations. The system takes a 400-point heart rate variability (HRV) sequence containing no AF episodes as the input and outputs whether the corresponding subject will experience AF episodes in the near future (i.e., 30 min). It first converts an input HRV sequence into four image matrices via extended Poincaré plots to capture inter- and intra-person features. Then, the system employs a convolutional neural network (CNN) to perform feature selection and classification based on the input image matrices. Some design issues of the system, including feature conversion and classifier structure, were formulated as a binary optimization problem, which was then solved via a genetic algorithm (GA). A numerical study involving 6085 400-point HRV sequences excerpted from three PhysioNet databases showed that the developed PAF prediction system achieved 87.9% and 87.2% accuracy on the validation and the testing datasets, respectively. The performance is competitive with that of the leading PAF prediction system in the literature, yet our system is much faster and more intensively tested. Furthermore, from the designed inter-person features, we found that PAF patients often possess lower (~60 beats/min) or higher (~100 beats/min) heart rates than non-PAF subjects. On the other hand, from the intra-person features, we observed that PAF patients often exhibit smaller variations (≤5 beats/min) in heart rate than non-PAF subjects, but they may experience short bursts of large heart rate changes sometimes, probably due to abnormal beats, such as premature atrial beats. The other findings warrant further investigations for their medical implications about the onset of PAF.

Opening the black box: interpretability of machine learning algorithms in electrocardiography

Recent studies have suggested that cardiac abnormalities can be detected from the electrocardiogram (ECG) using deep machine learning (DL) models. However, most DL algorithms lack interpretability, since they do not provide any justification for their decisions. In this study, we designed two new frameworks to interpret the classification results of DL algorithms trained for 12-lead ECG classification. The frameworks allow us to highlight not only the ECG samples that contributed most to the classification, but also which between the P-wave, QRS complex and T-wave, hereafter simply called 'waves', were the most relevant for the diagnosis. The frameworks were designed to be compatible with any DL model, including the ones already trained. The frameworks were tested on a selected Deep Neural Network, trained on a publicly available dataset, to automatically classify 24 cardiac abnormalities from 12-lead ECG signals. Experimental results showed that the frameworks were able to detect the most relevant ECG waves contributing to the classification. Often the network relied on portions of the ECG which are also considered by cardiologists to detect the same cardiac abnormalities, but this was not always the case. In conclusion, the proposed frameworks may unveil whether the network relies on features which are clinically significant for the detection of cardiac abnormalities from 12-lead ECG signals, thus increasing the trust in the DL models. This article is part of the theme issue 'Advanced computation in cardiovascular physiology: new challenges and opportunities'.

Symmetric Projection Attractor Reconstruction: Sex Differences in the ECG

Background: The electrocardiogram (ECG) is a key tool in patient management. Automated ECG analysis supports clinical decision-making, but traditional fiducial point identification discards much of the time-series data that captures the morphology of the whole waveform. Our Symmetric Projection Attractor Reconstruction (SPAR) method uses all the available data to provide a new visualization and quantification of the morphology and variability of any approximately periodic signal. We therefore applied SPAR to ECG signals to ascertain whether this more detailed investigation of ECG morphology adds clinical value. Methods: Our aim was to demonstrate the accuracy of the SPAR method in discriminating between two biologically distinct groups. As sex has been shown to influence the waveform appearance, we investigated sex differences in normal sinus rhythm ECGs. We applied the SPAR method to 9,007 10 second 12-lead ECG recordings from Physionet, which comprised; Dataset 1: 104 subjects (40% female), Dataset 2: 8,903 subjects (54% female). Results: SPAR showed clear visual differences between female and male ECGs (Dataset 1). A stacked machine learning model achieved a cross-validation sex classification accuracy of 86.3% (Dataset 2) and an unseen test accuracy of 91.3% (Dataset 1). The mid-precordial leads performed best in classification individually, but the highest overall accuracy was achieved with all 12 leads. Classification accuracy was highest for young adults and declined with older age. Conclusions: SPAR allows quantification of the morphology of the ECG without the need to identify conventional fiducial points, whilst utilizing of all the data reduces inadvertent bias. By intuitively re-visualizing signal morphology as two-dimensional images, SPAR accurately discriminated ECG sex differences in a small dataset. We extended the approach to a machine learning classification of sex for a larger dataset, and showed that the SPAR method provided a means of visualizing the similarities of subjects given the same classification. This proof-of-concept study therefore provided an implementation of SPAR using existing data and showed that subtle differences in the ECG can be amplified by the attractor. SPAR's supplementary analysis of ECG morphology may enhance conventional automated analysis in clinically important datasets, and improve patient stratification and risk management.

Robust R-Peak Detection in Low-Quality Holter ECGs using 1D Convolutional Neural Network

OBJECTIVE

Noise and low quality of ECG signals acquired from Holter or wearable devices deteriorate the accuracy and robustness of R-peak detection algorithms. This paper presents a generic and robust system for R-peak detection in Holter ECG signals. While many proposed algorithms have successfully addressed the problem of ECG R-peak detection, there is still a notable gap in the performance of these detectors on such low-quality ECG records.

METHODS

In this study, a novel implementation of the 1D Convolutional Neural Network (CNN) is used integrated with a verification model to reduce the number of false alarms. This CNN architecture consists of an encoder block and a corresponding decoder block followed by a sample-wise classification layer to construct the 1D segmentation map of R-peaks from the input ECG signal. Once the proposed model has been trained, it can solely be used to detect R-peaks possibly in a single channel ECG data stream quickly and accurately, or alternatively, such a solution can be conveniently employed for real-time monitoring on a lightweight portable device.

RESULTS

The model is tested on two open-access ECG databases: The China Physiological Signal Challenge (2020) database (CPSC-DB) with more than one million beats, and the commonly used MIT-BIH Arrhythmia Database (MIT-DB). Experimental results demonstrate that the proposed systematic approach achieves 99.30% F1-score, 99.69% recall, and 98.91% precision in CPSC-DB, which is the best R-peak detection performance ever achieved. Results also demonstrate similar or better performance than most competing algorithms on MIT-DB with 99.83% F1-score, 99.85% recall, and 99.82% precision.

SIGNIFICANCE

Compared to all competing methods, the proposed approach can reduce the false-positives and false-negatives in Holter ECG signals by more than 54% and 82%, respectively.

CONCLUSION

Finally, the simple and invariant nature of the parameters leads to a highly generic system and therefore applicable to any ECG dataset.

Modeling and Reproducing Textile Sensor Noise: Implications for Textile-Compatible Signal Processing Algorithms

Smart textiles provide an opportunity to simultaneously record various electrophysiological signals from the human body, such as ECG, in a non-invasive and continuous manner. Accurate processing of ECG signals recorded using textile sensors is challenging due to the very low signal-to-noise ratio (SNR). Signal processing algorithms that can extract ECG signal out of textile-based electrode recordings, despite low SNR are needed. Presently, there are no textile ECG datasets available to develop, test and validate these algorithms. In this paper we attempted to model textile ECG signals by adding the textile sensor noise to open access ECG signals. We employed the linear predictive coding method to model different features of this noise. By approximating the linear predictive coding residual signals using Kernel Density Estimation, an artificial textile ECG noise signal was generated by filtering the residual signal with the linear predictive coding coefficients. The obtained textile sensor noise was added to the MIT-BIH Arrhythmia Database (MITDB), thus creating Textile-like ECG dataset consisting of 108 channels (30 min each). Furthermore, a Python code for generating textile-like ECG signals with variable SNR was also made available online. Finally, to provide a benchmark for the performance of R-peak detection algorithms on textile ECG, the five common R-peak detection algorithms: Pan & Tompkins, improved Pan & Tompkins (in Biosppy), Hamilton, Engelse, and Khamis, were tested on textile-like MITDB. This work provides an approach to generating noisy textile ECG signals, and facilitating the development, testing, and evaluation of signal processing algorithms for textile ECGs.

Diagnostic Interpretation of Non-Uniformly Sampled Electrocardiogram

We present a set of three fundamental methods for electrocardiogram (ECG) diagnostic interpretation adapted to process non-uniformly sampled signal. The growing volume of ECGs recorded daily all over the world (roughly estimated to be 600 TB) and the expectance of long persistence of these data (on the order of 40 years) motivated us to challenge the feasibility of medical-grade diagnostics directly based on arbitrary non-uniform (i.e., storage-efficient) ECG representation. We used a refined time-independent QRS detection method based on a moving shape matching technique. We applied a graph data representation to quantify the similarity of asynchronously sampled heartbeats. Finally, we applied a correlation-based non-uniform to time-scale transform to get a multiresolution ECG representation on a regular dyadic grid and to find precise P, QRS and T wave delimitation points. The whole processing chain was implemented and tested with MIT-BIH Database (probably the most referenced cardiac database) and CSE Multilead Database (used for conformance testing of medical instruments) signals arbitrarily sampled accordingly to a perceptual model (set for variable sampling frequency of 100-500 Hz, compression ratio 3.1). The QRS detection shows an accuracy of 99.93% with false detection ratio of only 0.18%. The classification shows an accuracy of 99.27% for 14 most frequent MIT-BIH beat types and 99.37% according to AAMI beat labels. The wave delineation shows cumulative (i.e., sampling model and non-uniform processing) errors of: 9.7 ms for P wave duration, 3.4 ms for QRS, 6.7 ms for P-Q segment and 17.7 ms for Q-T segment, all the values being acceptable for medical-grade interpretive software.

Digital biomarkers and algorithms for detection of atrial fibrillation using surface electrocardiograms: A systematic review

AIMS

Automated detection of atrial fibrillation (AF) in continuous rhythm registrations is essential in order to prevent complications and optimize treatment of AF. Many algorithms have been developed to detect AF in surface electrocardiograms (ECGs) during the past few years. The aim of this systematic review is to gain more insight into these available classification methods by discussing previously used digital biomarkers and algorithms and make recommendations for future research.

METHODS

On the 14^th of September 2020, the PubMed database was searched for articles focusing on algorithms for AF detection in ECGs using the MeSH terms Atrial Fibrillation, Electrocardiography and Algorithms. Articles which solely focused on differentiation of types of rhythm disorders or prediction of AF termination were excluded.

RESULTS

The search resulted in 451 articles, of which 130 remained after full-text screening. Not only did the amount of research on methods for AF detection increase over the past years, but a trend towards more complex classification methods is observed. Furthermore, three different types of features can be distinguished: atrial features, ventricular features, and signal features. Although AF is an atrial disease, only 22% of the described methods use atrial features.

CONCLUSION

More and more studies focus on improving accuracy of classification methods for AF in ECGs. As a result, algorithms become increasingly complex and less well interpretable. Only a few studies focus on detecting atrial activity in the ECG. Developing innovative methods focusing on detection of atrial activity might provide accurate classifiers without compromising on transparency.

Automatic Detection of QRS Complexes Using Dual Channels Based on U-Net and Bidirectional Long Short-Term Memory

OBJECTIVE

Detecting changes in the QRS complexes in ECG signals is regarded as a straightforward, noninvasive, inexpensive, and preliminary diagnosis approach for evaluating the cardiac health of patients. Therefore, detecting QRS complexes in ECG signals must be accurate over short times. However, the reliability of automatic QRS detection is restricted by all kinds of noise and complex signal morphologies. The objective of this paper is to address automatic detection of QRS complexes.

METHODS

In this paper, we proposed a new algorithm for automatic detection of QRS complexes using dual channels based on U-Net and bidirectional long short-term memory. First, a proposed preprocessor with mean filtering and discrete wavelet transform was initially applied to remove different types of noise. Next the signal was transformed and annotations were relabeled. Finally, a method combining U-Net and bidirectional long short-term memory with dual channels was used for the automatic detection of QRS complexes.

RESULTS

The proposed algorithm was trained and tested using 44 ECG records from the MIT-BIH arrhythmia database and CPSC2019 dataset, which achieved 99.06% and 95.13% for sensitivity, 99.22% and 82.03% for positive predictivity, and 98.29% and 78.73% accuracy on the two datasets respectively.

CONCLUSION

Experimental results prove that the proposed method may be useful for automatic detection of QRS complex task.

SIGNIFICANCE

The proposed method not only has application potential for QRS complex detecting for large ECG data, but also can be extended to other medical signal research fields.

Arrhythmia classification from single-lead ECG signals using the inter-patient paradigm

BACKGROUND AND OBJECTIVES

Arrhythmia is a heart disease characterized by the change in the regularity of the heartbeat. Since this disorder can occur sporadically, Holter devices are used for continuous long-term monitoring of the subject's electrocardiogram (ECG). In this process, a large volume of data is generated. Consequently, the use of an automated system for detecting arrhythmias is highly desirable. In this work, an automated system for classifying arrhythmias using single-lead ECG signals is proposed.

METHODS

The proposed system uses a combination of three groups of features: RR intervals, signal morphology, and higher-order statistics. To validate the method, the MIT-BIH database was employed using the inter-patient paradigm. Besides, the robustness of the system against segmentation errors was tested by adding jitter to the R-wave positions given by the MIT-BIH database. Additionally, each group of features had its robustness against segmentation error tested as well.

RESULTS

The experimental results of the proposed classification system with jitter show that the sensitivities for the classes N, S, and V are 93.7, 89.7, and 87.9, respectively. Also, the corresponding positive predictive values are 99.2, 36.8, and 93.9, respectively.

CONCLUSIONS

The proposed method was able to outperform several state-of-the-art methods, even though the R-wave position was synthetically corrupted by added jitter. The obtained results show that our approach can be employed in real scenarios where segmentation errors and the inter-patient paradigm are present.

The hidden waves in the ECG uncovered revealing a sound automated interpretation method

A novel approach for analysing cardiac rhythm data is presented in this paper. Heartbeats are decomposed into the five fundamental P, Q, R, S and T waves plus an error term to account for artifacts in the data which provides a meaningful, physical interpretation of the heart's electric system. The morphology of each wave is concisely described using four parameters that allow all the different patterns in heartbeats to be characterized and thus differentiated This multi-purpose approach solves such questions as the extraction of interpretable features, the detection of the fiducial marks of the fundamental waves, or the generation of synthetic data and the denoising of signals. Yet the greatest benefit from this new discovery will be the automatic diagnosis of heart anomalies as well as other clinical uses with great advantages compared to the rigid, vulnerable and black box machine learning procedures, widely used in medical devices. The paper shows the enormous potential of the method in practice; specifically, the capability to discriminate subjects, characterize morphologies and detect the fiducial marks (reference points) are validated numerically using simulated and real data, thus proving that it outperforms its competitors.

RPnet: A Deep Learning approach for robust R Peak detection in noisy ECG

Automatic detection of R-peaks in an Electrocardiogram signal is crucial in a multitude of applications including Heart Rate Variability (HRV) analysis and Cardio Vascular Disease(CVD) diagnosis. Although there have been numerous approaches that have successfully addressed the problem, there has been a notable dip in the performance of these existing detectors on ECG episodes that contain noise and HRV Irregulates. On the other hand, Deep Learning(DL) based methods have shown to be adept at modelling data that contain noise. In image to image translation, Unet is the fundamental block in many of the networks. In this work, a novel application of the Unet combined with Inception and Residual blocks is proposed to perform the extraction of R-peaks from an ECG. Furthermore, the problem formulation also robustly deals with issues of variability and sparsity of ECG R-peaks. The proposed network was trained on a database containing ECG episodes that have CVD and was tested against three traditional ECG detectors on a validation set. The model achieved an F1 score of 0.9837, which is a substantial improvement over the other beat detectors. Furthermore, the model was also evaluated on three other databases. The proposed network achieved high F1 scores across all datasets which established its generalizing capacity. Additionally, a thorough analysis of the model's performance in the presence of different levels of noise was carried out.

Suppressing the Influence of Ectopic Beats by Applying a Physical Threshold-Based Sample Entropy

Sample entropy (SampEn) is widely used for electrocardiogram (ECG) signal analysis to quantify the inherent complexity or regularity of RR interval time series (i.e., heart rate variability (HRV)), with the hypothesis that RR interval time series in pathological conditions output lower SampEn values. However, ectopic beats can significantly influence the entropy values, resulting in difficulty in distinguishing the pathological situation from normal situations. Although a theoretical operation is to exclude the ectopic intervals during HRV analysis, it is not easy to identify all of them in practice, especially for the dynamic ECG signal. Thus, it is important to suppress the influence of ectopic beats on entropy results, i.e., to improve the robustness and stability of entropy measurement for ectopic beats-inserted RR interval time series. In this study, we introduced a physical threshold-based SampEn method, and tested its ability to suppress the influence of ectopic beats for HRV analysis. An experiment on the PhysioNet/MIT RR Interval Databases showed that the SampEn use physical meaning threshold has better performance not only for different data types (normal sinus rhythm (NSR) or congestive heart failure (CHF) recordings), but also for different types of ectopic beat (atrial beats, ventricular beats or both), indicating that using a physical meaning threshold makes SampEn become more consistent and stable.

Multiple Physiological Signals Fusion Techniques for Improving Heartbeat Detection: A Review

This paper presents a review of the techniques found in the literature that aim to achieve a robust heartbeat detection from fusing multi-modal physiological signals (e.g., electrocardiogram (ECG), blood pressure (BP), artificial blood pressure (ABP), stroke volume (SV), photoplethysmogram (PPG), electroencephalogram (EEG), electromyogram (EMG), and electrooculogram (EOG), among others). Techniques typically employ ECG, BP, and ABP, of which usage has been shown to obtain the best performance under challenging conditions. SV, PPG, EMG, EEG, and EOG signals can help increase performance when included within the fusion. Filtering, signal normalization, and resampling are common preprocessing steps. Delay correction between the heartbeats obtained over some of the physiological signals must also be considered, and signal-quality assessment to retain the best signal/s must be considered as well. Fusion is usually accomplished by exploiting regularities in the RR intervals; by selecting the most promising signal for the detection at every moment; by a voting process; or by performing simultaneous detection and fusion using Bayesian techniques, hidden Markov models, or neural networks. Based on the results of the review, guidelines to facilitate future comparison of the performance of the different proposals are given and promising future lines of research are pointed out.

Enabling Heart Self-Monitoring for All and for AAL-Portable Device within a Complete Telemedicine System

During the last decades there has been a rapidly growing elderly population and the number of patients with chronic heart-related diseases has exploded. Many of them (such as those with congestive heart failure or some types of arrhythmias) require close medical supervision, thus imposing a big burden on healthcare costs in most western economies. Specifically, continuous or frequent Arterial Blood Pressure (ABP) and electrocardiogram (ECG) monitoring are important tools in the follow-up of many of these patients. In this work, we present a novel remote non-ambulatory and clinically validated heart self-monitoring system, which allows ABP and ECG monitoring to effectively identify clinically relevant arrhythmias. The system integrates digital transmission of the ECG and tensiometer measurements, within a patient-comfortable support, easy to recharge and with a multi-function software, all of them aiming to adapt for elderly people. The main novelty is that both physiological variables (ABP and ECG) are simultaneously measured in an ambulatory environment, which to our best knowledge is not readily available in the clinical market. Different processing techniques were implemented to analyze the heart rhythm, including pause detection, rhythm alterations and atrial fibrillation, hence allowing early detection of these diseases. Our results achieved clinical quality both for in-lab hardware testing and for ambulatory scenario validations. The proposed active assisted living (AAL) Sensor-based system is an end-to-end multidisciplinary system, fully connected to a platform and tested by the clinical team from beginning to end.

An Improved Sliding Window Area Method for T Wave Detection

BACKGROUND

The T wave represents ECG repolarization, whose detection is required during myocardial ischemia, and the first significant change in the ECG signal is being observed in the ST segment followed by changes in other waves like P wave and QRS complex. To offer guidance in clinical diagnosis, decision-making, and daily mobile ECG monitoring, the T wave needs to be detected firstly. Recently, the sliding area-based method has received an increasing amount of attention due to its robustness and low computational burden. However, the parameter setting of the search window's boundaries in this method is not adaptive. Therefore, in this study, we proposed an improved sliding window area method with more adaptive parameter setting for T wave detection.

METHODS

Firstly, k-means clustering was used in the annotated MIT QT database to generate three piecewise functions for delineating the relationship between the RR interval and the interval from the R peak to the T wave onset and that between the RR interval and the interval from the R peak to the T wave offset. Then, the grid search technique combined with 5-fold cross validation was used to select the suitable parameters' combination for the sliding window area method.

RESULTS

With respect to onset detection in the QT database, F1 improved from 54.70% to 70.46% and 54.05% to 72.94% for the first and second electrocardiogram (ECG) channels, respectively. For offset detection, F1 also improved in both channels as it did in the European ST-T database.

CONCLUSIONS

F1 results from the improved algorithm version were higher than those from the traditional method, indicating a potentially useful application for the proposed method in ECG monitoring.