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

Visual interpretation of deep learning model in ECG classification: A comprehensive evaluation of feature attribution methods

Feature attribution methods can visually highlight specific input regions containing influential aspects affecting a deep learning model's prediction. Recently, the use of feature attribution methods in electrocardiogram (ECG) classification has been sharply increasing, as they assist clinicians in understanding the model's decision-making process and assessing the model's reliability. However, a careful study to identify suitable methods for ECG datasets has been lacking, leading researchers to select methods without a thorough understanding of their appropriateness. In this work, we conduct a large-scale assessment by considering eleven popular feature attribution methods across five large ECG datasets using a model based on the ResNet-18 architecture. Our experiments include both automatic evaluations and human evaluations. Annotated datasets were utilized for automatic evaluations and three cardiac experts were involved for human evaluations. We found that Guided Grad-CAM, particularly when its absolute values are utilized, achieves the best performance. When Guided Grad-CAM was utilized as the feature attribution method, cardiac experts confirmed that it can identify diagnostically relevant electrophysiological characteristics, although its effectiveness varied across the 17 different diagnoses that we have investigated.

Heart rate variability for neuro-prognostication after CA: Insight from the Parisian registry

BACKGROUND

Hypoxic ischemic brain injury (HIBI) induced by cardiac arrest (CA) seems to predominate in cortical areas and to a lesser extent in the brainstem. These regions play key roles in modulating the activity of the autonomic nervous system (ANS), that can be assessed through analyses of heart rate variability (HRV). The objective was to evaluate the prognostic value of various HRV parameters to predict neurological outcome after CA.

METHODS

Retrospective monocentric study assessing the prognostic value of HRV markers and their association with HIBI severity. Patients admitted for CA who underwent EEG for persistent coma after CA were included. HRV markers were computed from 5 min signal of the ECG lead of the EEG recording. HRV indices were calculated in the time-, frequency-, and non-linear domains. Frequency-domain analyses differentiated very low frequency (VLF 0.003-0.04 Hz), low frequency (LF 0.04-0.15 Hz), high frequency (HF 0.15-0.4 Hz), and LF/HF ratio. HRV indices were compared to other prognostic markers: pupillary light reflex, EEG, N20 on somatosensory evoked potentials (SSEP) and biomarkers (neuron specific enolase-NSE). Neurological outcome at 3 months was defined as unfavorable in case of best CPC 3-4-5.

RESULTS

Between 2007 and 2021, 199 patients were included. Patients were predominantly male (64%), with a median age of 60 [48.9-71.7] years. 76% were out-of-hospital CA, and 30% had an initial shockable rhythm. Neurological outcome was unfavorable in 73%. Compared to poor outcome, patients with a good outcome had higher VLF (0.21 vs 0.09 ms2/Hz, p < 0.01), LF (0.07 vs 0.04 ms2/Hz, p = 0.003), and higher LF/HF ratio (2.01 vs 1.01, p = 0.008). Several non-linear domain indices were also higher in the good outcome group, such as SD2 (15.1 vs 10.2, p = 0.016) and DFA α1 (1.03 vs 0.78, p = 0.002). These indices also differed depending on the severity of EEG pattern and abolition of pupillary light reflex. These time-frequency and non-linear domains HRV parameters were predictive of poor neurological outcome, with high specificity despite a low sensitivity.

CONCLUSION

In comatose patients after CA, some HRV markers appear to be associated with unfavorable outcome, EEG severity and PLR abolition, although the sensitivity of these HRV markers remains limited.

QRS detection in single-lead, telehealth electrocardiogram signals: Benchmarking open-source algorithms

BACKGROUND AND OBJECTIVES

A key step in electrocardiogram (ECG) analysis is the detection of QRS complexes, particularly for arrhythmia detection. Telehealth ECGs present a new challenge for automated analysis as they are noisier than traditional clinical ECGs. The aim of this study was to identify the best-performing open-source QRS detector for use with telehealth ECGs.

METHODS

The performance of 18 open-source QRS detectors was assessed on six datasets. These included four datasets of ECGs collected under supervision, and two datasets of telehealth ECGs collected without clinical supervision. The telehealth ECGs, consisting of single-lead ECGs recorded between the hands, included a novel dataset of 479 ECGs collected in the SAFER study of screening for atrial fibrillation (AF). Performance was assessed against manual annotations.

RESULTS

A total of 12 QRS detectors performed well on ECGs collected under clinical supervision (F1 score ≥0.96). However, fewer performed well on telehealth ECGs: five performed well on the TELE ECG Database; six performed well on high-quality SAFER data; and performance was poorer on low-quality SAFER data (three QRS detectors achieved F1 of 0.78-0.84). The presence of AF had little impact on performance.

CONCLUSIONS

The Neurokit and University of New South Wales QRS detectors performed best in this study. These performed sufficiently well on high-quality telehealth ECGs, but not on low-quality ECGs. This demonstrates the need to handle low-quality ECGs appropriately to ensure only ECGs which can be accurately analysed are used for clinical decision making.

Automated cardiac arrhythmia detection techniques: a comprehensive review for prospective approach

Abnormal cardiac functionality produces irregular heart rhythms which are commonly known as arrhythmias. In some conditions, arrhythmias are treated as very dangerous which may lead to sudden cardiac arrest. The incidence and prevalence of cardiac anomalies seeks early detection of arrhythmias using automated classification techniques. In the past, numerous automated arrhythmia detection techniques have been developed that are based on electrocardiogram (ECG) signal analysis. Focusing on the prospective research in this field, this article reports a comprehensive review of existing techniques that are obtained using search engines such as IEEE explore, Google scholar and science direct. Based on the review, the existing techniques are broadly categorized into two types: machine-learning and deep-learning-based techniques. In this study, it is noticed that the performance of the machine-learning-based arrhythmia detection techniques depend on pre-processing of ECG signal, R-peaks detection, features extraction and classification tools while the deep-learning-based techniques do not require the features extraction step. Generally, the existing techniques utilize Massachusetts Institute of Technology-Beth Israel Hospital arrhythmia database to evaluate the classification performance. The classification performance of automated techniques also depends on ECG data used for training and testing of the classifier. It is expected that the performance should be evaluated using a variety of ECG signals including the cases of inter-patient and intra-patient paradigm. The existing techniques also require to deal with the class-imbalance problem. In addition to this, a specific partition-ratio between training and testing datasets should be maintained for fair comparison of performance of different techniques.

Aberrant brain-heart coupling is associated with the severity of post cardiac arrest brain injury

OBJECTIVE

To investigate autonomic nervous system activity measured by brain-heart interactions in comatose patients after cardiac arrest in relation to the severity and prognosis of hypoxic-ischemic brain injury.

METHODS

Strength and complexity of bidirectional interactions between EEG frequency bands (delta, theta, and alpha) and ECG heart rate variability frequency bands (low frequency, LF and high frequency, HF) were computed using a synthetic data generation model. Primary outcome was the severity of brain injury, assessed by (i) standardized qualitative EEG classification, (ii) somatosensory evoked potentials (N20), and (iii) neuron-specific enolase levels. Secondary outcome was the 3-month neurological status, assessed by the Cerebral Performance Category score [good (1-2) vs. poor outcome (3-4-5)].

RESULTS

Between January 2007 and July 2021, 181 patients were admitted to ICU for a resuscitated cardiac arrest. Poor neurological outcome was observed in 134 patients (74%). Qualitative EEG patterns suggesting high severity were associated with decreased LF/HF. Severity of EEG changes were proportional to higher absolute values of brain-to-heart coupling strength (p < 0.02 for all brain-to-heart frequencies) and lower values of alpha-to-HF complexity (p = 0.049). Brain-to-heart coupling strength was significantly higher in patients with bilateral absent N20 and correlated with neuron-specific enolase levels at Day 3. This aberrant brain-to-heart coupling (increased strength and decreased complexity) was also associated with 3-month poor neurological outcome.

INTERPRETATION

Our results suggest that autonomic dysfunctions may well represent hypoxic-ischemic brain injury post cardiac arrest pathophysiology. These results open avenues for integrative monitoring of autonomic functioning in critical care patients.

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.

A Review on Atrial Fibrillation Detection From Ambulatory ECG

Atrial fibrillation (AF) is a prevalent clinical arrhythmia disease and is an important cause of stroke, heart failure, and sudden death. Due to the insidious onset and no obvious clinical symptoms of AF, the status of AF diagnosis and treatment is not optimal. Early AF screening or detection is essential. Internet of Things (IoT) and artificial intelligence (AI) technologies have driven the development of wearable electrocardiograph (ECG) devices used for health monitoring, which are an effective means of AF detection. The main challenges of AF analysis using ambulatory ECG include ECG signal quality assessment to select available ECG, the robust and accurate detection of QRS complex waves to monitor heart rate, and AF identification under the interference of abnormal ECG rhythm. Through ambulatory ECG measurement and intelligent detection technology, the probability of postoperative recurrence of AF can be reduced, and personalized treatment and management of patients with AF can be realized. This work describes the status of AF monitoring technology in terms of devices, algorithms, clinical applications, and future directions.

Tracking Cancer: Exploring Heart Rate Variability Patterns by Cancer Location and Progression

Reduced heart rate variability (HRV) is an autonomic nervous system (ANS) response that may indicate dysfunction in the human body. Consistent evidence shows cancer patients elicit lower HRV; however, only select cancer locations were previously evaluated. Thus, the aim of the current study was to explore HRV patterns in patients diagnosed with and in varying stages of the most prevalent cancers. At a single tertiary academic medical center, 798 patients were recruited. HRV was measured via an armband monitor (Warfighter MonitorTM, Tiger Tech Solutions, Inc., Miami, FL, USA) equipped with electrocardiographic capabilities and was recorded for 5 to 7 min with patients seated in an upright position. Three time-domain metrics were calculated: SDNN (standard deviation of the NN interval), rMSSD (the root mean square of successive differences of NN intervals), and the percentage of time in which the change in successive NN intervals exceeds 50ms within a measurement (pNN50). Of the 798 patients, 399 were diagnosed with cancer. Cancer diagnoses were obtained via medical records one week following the measurement. Analysis of variance models were performed comparing the HRV patterns between different cancers, cancer stages (I-IV), and demographic strata. A total of 85% of the cancer patients had breast, gastrointestinal, genitourinary, or respiratory cancer. The cancer patients were compared to a control non-cancer patient population with similar patient size and distributions for sex, age, body mass index, and co-morbidities. For all HRV metrics, non-cancer patients exhibited significantly higher rMSSDs (11.1 to 13.9 ms, p < 0.0001), SDNNs (22.8 to 27.7 ms, p < 0.0001), and pNN50s (6.2 to 8.1%, p < 0.0001) compared to stage I or II cancer patients. This significant trend was consistently observed across each cancer location. Similarly, compared to patients with stage III or IV cancer, non-cancer patients possessed lower HRs (-11.8 to -14.0 bpm, p < 0.0001) and higher rMSSDs (+31.7 to +32.8 ms, p < 0.0001), SDNNs (+45.2 to +45.8 ms), p < 0.0001, and pNN50s (19.2 to 21.6%, p < 0.0001). The HR and HRV patterns observed did not significantly differ between cancer locations (p = 0.96 to 1.00). The depressed HRVs observed uniformly across the most prevalent cancer locations and stages appeared to occur independent of patients' co-morbidities. This finding highlights the potentially effective use of HRV as a non-invasive tool for determining common cancer locations and their respective stages. More studies are needed to delineate the HRV patterns across different ages, between sexes and race/ethnic groups.

Improved diagnostic performance of insertable cardiac monitors by an artificial intelligence-based algorithm

AIMS

The increasing use of insertable cardiac monitors (ICM) produces a high rate of false positive (FP) diagnoses. Their verification results in a high workload for caregivers. We evaluated the performance of an artificial intelligence (AI)-based ILR-ECG Analyzer™ (ILR-ECG-A). This machine-learning algorithm reclassifies ICM-transmitted events to minimize the rate of FP diagnoses, while preserving device sensitivity.

METHODS AND RESULTS

We selected 546 recipients of ICM followed by the Implicity™ monitoring platform. To avoid clusterization, a single episode per ICM abnormal diagnosis (e.g. asystole, bradycardia, atrial tachycardia (AT)/atrial fibrillation (AF), ventricular tachycardia, artefact) was selected per patient, and analyzed by the ILR-ECG-A, applying the same diagnoses as the ICM. All episodes were reviewed by an adjudication committee (AC) and the results were compared. Among 879 episodes classified as abnormal by the ICM, 80 (9.1%) were adjudicated as 'Artefacts', 283 (32.2%) as FP, and 516 (58.7%) as 'abnormal' by the AC. The algorithm reclassified 215 of the 283 FP as normal (76.0%), and confirmed 509 of the 516 episodes as abnormal (98.6%). Seven undiagnosed false negatives were adjudicated as AT or non-specific abnormality. The overall diagnostic specificity was 76.0% and the sensitivity was 98.6%.

CONCLUSION

The new AI-based ILR-ECG-A lowered the rate of FP ICM diagnoses significantly while retaining a > 98% sensitivity. This will likely alleviate considerably the clinical burden represented by the review of ICM events.

Fast Parabolic Fitting: An R-Peak Detection Algorithm for Wearable ECG Devices

Heart diseases rank among the most fatal health concerns globally, with the majority being preventable through early diagnosis and effective treatment. Electrocardiogram (ECG) analysis is critical in detecting heart diseases, as it captures the heart's electrical activities. For continuous monitoring, wearable electrocardiographic devices must ensure user comfort over extended periods, typically 24 to 48 h. These devices demand specialized algorithms with low computational complexity to accommodate memory and power consumption constraints. One of the most crucial aspects of ECG signals is accurately detecting heartbeat intervals, specifically the R peaks. In this study, we introduce a novel algorithm designed for wearable devices, offering two primary attributes: robustness against noise and low computational complexity. Our algorithm entails fitting a least-squares parabola to the ECG signal and adaptively shaping it as it sweeps through the signal. Notably, our proposed algorithm eliminates the need for band-pass filters, which can inadvertently smooth the R peaks, making them more challenging to identify. We compared the algorithm's performance using two extensive databases: the meta-database QT database and the BIH-MIT database. Importantly, our method does not necessitate the precise localization of the ECG signal's isoelectric line, contributing to its low computational complexity. In the analysis of the QT database, our algorithm demonstrated a substantial advantage over the classical Pan-Tompkins algorithm and maintained competitiveness with state-of-the-art approaches. In the case of the BIH-MIT database, the performance results were more conservative; they continued to underscore the real-world utility of our algorithm in clinical contexts.

End-to-End Premature Ventricular Contraction Detection Using Deep Neural Networks

In Holter monitoring, the precise detection of standard heartbeats and ventricular premature contractions (PVCs) is paramount for accurate cardiac rhythm assessment. This study introduces a novel application of the 1D U-Net neural network architecture with the aim of enhancing PVC detection in Holter recordings. Training data comprised the Icentia 11k and INCART DB datasets, as well as our custom dataset. The model's efficacy was subsequently validated against traditional Holter analysis methodologies across multiple databases, including AHA DB, MIT 11 DB, and NST, as well as another custom dataset that was specifically compiled by the authors encompassing challenging real-world examples. The results underscore the 1D U-Net model's prowess in QRS complex detection, achieving near-perfect balanced accuracy scores across all databases. PVC detection exhibited variability, with balanced accuracy scores ranging from 0.909 to 0.986. Despite some databases, like the AHA DB, showcasing lower sensitivity metrics, their robust, balanced accuracy accentuates the model's equitable performance in discerning both false positives and false negatives. In conclusion, while the 1D U-Net architecture is a formidable tool for QRS detection, there's a clear avenue for further refinement in its PVC detection capability, given the inherent complexities and noise challenges in real-world PVC occurrences.

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.

Automated electrocardiogram signal quality assessment based on Fourier analysis and template matching

We developed and tested a novel template matching approach for signal quality assessment on electrocardiogram (ECG) data. A computational method was developed that uses a sinusoidal approximation to the QRS complex to generate a correlation value at every point of an ECG. The strength of this correlation can be numerically adapted into a 'score' for each segment of an ECG, which can be used to stratify signal quality. The algorithm was tested on lead II ECGs of intensive care unit (ICU) patients admitted to the Mount Sinai Hospital (MSH) from January to July 2020 and on records from the MIT BIH arrhythmia database. The algorithm was found to be 98.9% specific and 99% sensitive on test data from the MSH ICU patients. The routine performs in linear O(n) time and occupies O(1) heap space in runtime. This approach can be used to lower the burden of pre-processing in ECG signal analysis. Given its runtime (O(n)) and memory (O(1)) complexity, there are potential applications for signal quality stratification and arrhythmia detection in wearable devices or smartphones.

ECG signal feature extraction trends in methods and applications

Signal analysis is a domain which is an amalgamation of different processes coming together to form robust pipelines for the automation of data analysis. When applied to the medical world, physiological signals are used. It is becoming increasingly common in today's day and age to be working with very large datasets, on the scale of having thousands of features. This is largely due to the fact that the acquisition of biomedical signals can be taken over multi-hour timeframes, which is another challenge to solve in and of itself. This paper will focus on the electrocardiogram (ECG) signal specifically, and common feature extraction techniques used for digital health and artificial intelligence (AI) applications. Feature extraction is a vital step of biomedical signal analysis. The basic goal of feature extraction is for signal dimensionality reduction and data compaction. In simple terms, this would allow one to represent data with a smaller subset of features; these features could then later be leveraged to be used more efficiently for machine learning and deep learning models for applications, such as classification, detection, and automated applications. In addition, the redundant data in the overall dataset is filtered out as the data is reduced during feature extraction. In this review, we cover ECG signal processing and feature extraction in the time domain, frequency domain, time-frequency domain, decomposition, and sparse domain. We also provide pseudocode for the methods discussed so that they can be replicated by practitioners and researchers in their specific areas of biomedical work. Furthermore, we discuss deep features, and machine learning integration, to complete the overall pipeline design for signal analysis. Finally, we discuss future work that can be innovated upon in the feature extraction domain for ECG signal analysis.

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.

A training pipeline of an arrhythmia classifier for atrial fibrillation detection using Photoplethysmography signal

Photoplethysmography (PPG) signal is potentially suitable in atrial fibrillation (AF) detection for its convenience in use and similarity in physiological origin to electrocardiogram (ECG). There are a few preceding studies that have shown the possibility of using the peak-to-peak interval of the PPG signal (PPIp) in AF detection. However, as a generalized model, the accuracy of an AF detector should be pursued on the one hand; on the other hand, its generalizability should be paid attention to in view of the individual differences in PPG manifestation of even the same arrhythmia and the existence of sub-types. Moreover, a binary classifier for atrial fibrillation and normal sinus rhythm is not convincing enough for the similarity between AF and ectopic beats. In this study, we project the atrial fibrillation detection as a multiple-class classification and try to propose a training pipeline that is advantageous both to the accuracy and generalizability of the classifier by designing and determining the configurable options of the pipeline, in terms of input format, deep learning model (with hyperparameter optimization), and scheme of transfer learning. With a rigorous comparison of the possible combinations of the configurable components in the pipeline, we confirmed that first-order difference of heartbeat sequence as the input format, a 2-layer CNN-1-layer Transformer hybrid^R model as the learning model and the whole model fine-tuning as the implementing scheme of transfer learning is the best combination for the pipeline (F1 value: 0.80, overall accuracy: 0.87)^R.

A 2.2 nW Analog Electrocardiogram Processor based on Stochastic Resonance Achieving a 99.94% QRS Complex Detection Sensitivity

This paper presents an ultra-low power electrocardiogram (ECG) processor that can detect QRS-waves in real time as the data streams in. The processor performs out-of-band noise suppression via a linear filter, and in-band noise suppression via a nonlinear filter. The nonlinear filter also enhances the QRS-waves by facilitating stochastic resonance. The processor identifies the QRS-waves on noise-suppressed and enhanced recordings using a constant threshold detector. For energy-efficiency and compactness, the processor exploits current-mode analog signal processing techniques, which significantly reduces the design complexity when implementing the second-order dynamics of the nonlinear filter. The processor is designed and implemented in TSMC 65 nm CMOS technology. In terms of detection performance, the processor achieves an average F1 = 99.88% over the MIT-BIH Arrhythmia database and outperforms all previous ultra-low power ECG processors. The processor is the first that is validated against noisy ECG recordings of MIT-BIH NST and TELE databases, where it achieves better detection performances than most digital algorithms run on digital platforms. The design has a footprint of 0.08 mm² and dissipates 2.2 nW when supplied by a single 1V supply, making it the first ultra-low power and real-time processor that facilitates stochastic resonance.

Heart Rate Variability in Children with Moderate and Severe Traumatic Brain Injury: A Prospective Observational Study

The aim of this study was to assess the feasibility of continuous monitoring of heart rate variability (HRV) in children with traumatic brain injury (TBI) hospitalized in a pediatric intensive care unit (PICU) and collect preliminary data on the association between HRV, neurological outcome, and complications. This is a prospective observational cohort study in a tertiary academic PICU. Children admitted to the PICU ≤24 hours after moderate or severe TBI were included in the study. Children suspected of being brain dead at PICU entry or with a pacemaker were excluded. Children underwent continuous monitoring of electrocardiographic (ECG) waveforms over 7 days post-TBI. HRV analysis was performed retrospectively, using a standardized, validated HRV analysis software (CIMVA). The occurrence of medical complications (“event”: intracranial hypertension, cerebral hypoperfusion, seizure, and cardiac arrest) was prospectively documented. Outcome of children 6 months post-TBI was assessed using the Glasgow Outcome Scale – Extended Pediatric (GOS-E Peds). Fifteen patients were included over a 20-month period. Thirteen patients had ECG recordings available and 4 had >20% of missing ECG data. When ECG was available, HRV calculation was feasible (average 88%; range 70–97%). Significant decrease in overall HRV coefficient of variation and Poincaré SD2 (p < 0.05) at 6 hours post–PICU admission was associated with an unfavorable outcome (defined as GOS-E Peds ≥ 3, or a deterioration of ≥2 points over baseline score). Several HRV metrics exhibited significant and nonsignificant variation in HRV during event. This study demonstrates that it is feasible to monitor HRV in the PICU provided ECG data are available; however, missing ECG data are not uncommon. These preliminary data suggest that altered HRV is associated with unfavorable neurological outcome and in-hospital medical complications. Larger prospective studies are needed to confirm these findings and to explore if HRV offers reliable and clinically useful prediction data that may help clinical decision making.

QRS detection and classification in Holter ECG data in one inference step

While various QRS detection and classification methods were developed in the past, the Holter ECG data acquired during daily activities by wearable devices represent new challenges such as increased noise and artefacts due to patient movements. Here, we present a deep-learning model to detect and classify QRS complexes in single-lead Holter ECG. We introduce a novel approach, delivering QRS detection and classification in one inference step. We used a private dataset (12,111 Holter ECG recordings, length of 30 s) for training, validation, and testing the method. Twelve public databases were used to further test method performance. We built a software tool to rapidly annotate QRS complexes in a private dataset, and we annotated 619,681 QRS complexes. The standardised and down-sampled ECG signal forms a 30-s long input for the deep-learning model. The model consists of five ResNet blocks and a gated recurrent unit layer. The model's output is a 30-s long 4-channel probability vector (no-QRS, normal QRS, premature ventricular contraction, premature atrial contraction). Output probabilities are post-processed to receive predicted QRS annotation marks. For the QRS detection task, the proposed method achieved the F1 score of 0.99 on the private test set. An overall mean F1 cross-database score through twelve external public databases was 0.96 ± 0.06. In terms of QRS classification, the presented method showed micro and macro F1 scores of 0.96 and 0.74 on the private test set, respectively. Cross-database results using four external public datasets showed micro and macro F1 scores of 0.95 ± 0.03 and 0.73 ± 0.06, respectively. Presented results showed that QRS detection and classification could be reliably computed in one inference step. The cross-database tests showed higher overall QRS detection performance than any of compared methods.

Electromyography Parameter Variations with Electrocardiography Noise

Electromyograms (EMG signals) may be contaminated by electrocardiographic (ECG) signals that cannot be easily separated with traditional filters, because both signals have some overlapping spectral components. Therefore, the first challenge encountered in signal processing is to extract the ECG noise from the EMG signal. In this study, the EMG, mixed with different degrees of noise (ECG), is simulated to investigate the variations of the EMG features. Simulated data were derived from the MIT-BIH Noise Stress Test (NSTD) Database. Two EMG and four ECG data were composed with four EMG/ECG SNR to 32 simulated signals. Following Pan-Tompkins R-peak detection, four ECG removal methods were used to remove ECG with different compensation algorithms to obtain the denoised EMG signal. A total of 13 time-domain and four frequency-domain EMG features were calculated from the denoised EMG. In addition, the similarity of denoised EMG features compared to clean EMG was also evaluated. Our results showed that with the ratio EMG/ECG SNR = 10 and 20, the ECG can be almost ignored, and the similarity of EMG features is close to 1. When EMG/ECG SNR = 1 and 2, there is a large variation of EMG features. The results of our simulation study would be beneficial for understanding the variations of EMG features upon the different EMG/ECG SNR.

Peak Detection and HRV Feature Evaluation on ECG and PPG Signals

Heart Rate Variability (HRV) evaluates the autonomic nervous system regulation and can be used as a monitoring tool in conditions such as cardiovascular diseases, neuropathies and sleep staging. It can be extracted from the electrocardiogram (ECG) and the photoplethysmogram (PPG) signals. Typically, the HRV is obtained from the ECG processing. Being the PPG sensor widely used in clinical setups for physiological parameters monitoring such as blood oxygenation and ventilatory rate, the question arises regarding the PPG adequacy for HRV extraction. There is not a consensus regarding the PPG being able to replace the ECG in the HRV estimation. This work aims to be a contribution to this research area by comparing the HRV estimation obtained from simultaneously acquired ECG and PPG signals from forty subjects. A peak detection method is herein introduced based on the Hilbert transform: Hilbert Double Envelope Method (HDEM). Two other peak detector methods were also evaluated: Pan-Tompkins and Wavelet-based. HRV parameters for time, frequency and the non-linear domain were calculated for each algorithm and the Pearson correlation, T-test and RMSE were evaluated. The HDEM algorithm showed the best overall results with a sensitivity of 99.07% and 99.45% for the ECG and the PPG signals, respectively. For this algorithm, a high correlation and no significant differences were found between HRV features and the gold standard, for the ECG and PPG signals. The results show that the PPG is a suitable alternative to the ECG for HRV feature extraction.

A Stochastic Resonance Electrocardiogram Enhancement Algorithm for Robust QRS Detection

This study presents a new QRS detection algorithm making use of the background noise that is inevitably present in electrocardiogram (ECG) recordings. The algorithm suppresses noise, enhances the QRS-waves, and applies a threshold for QRS detection. Noise suppression and QRS enhancement are performed by a band-pass filter stage followed by a nonlinear stage based on the interaction of a particle inside an underdamped monostable potential well. The nonlinear stage maximizes the output when there is a QRS-wave and minimizes the output otherwise. One of the instruments that the nonlinear stage uses to enhance the QRS-waves is stochastic resonance, where the output is maximized for a non-zero intensity background noise. In terms of QRS-wave detection F1 score, which ranges from 98.87% to 99.99% on four major benchmarking databases (MIT-BIH Arrhythmia, QT, European ST-T, and MIT-BIH Noise Stress Test), the algorithm outperforms all existing ECG processing algorithms. The study, for the first time, demonstrates QRS-enhancement by facilitating stochastic resonance while suppressing in-band noise of ECG signals. Detecting QRS-waves as the ECG data streams, having a complexity of O(n), and not requiring any training data make the algorithm convenient for real-time ECG monitoring applications with limited computational resources.

ECGAssess: A Python-Based Toolbox to Assess ECG Lead Signal Quality

Electrocardiography (ECG) is the method most often used to diagnose cardiovascular diseases. To obtain a high-quality recording, the person conducting an ECG must be a trained expert. When these experts are not available, this important diagnostic tool cannot be used, consequently impacting the quality of healthcare. To avoid this problem, it must be possible for untrained healthcare professionals to record diagnostically useful ECGs so they can send the recordings to experts for diagnosis. The ECGAssess Python-based toolbox developed in this study provides feedback regarding whether ECG signals are of adequate quality. Each lead of the 12-lead recordings was classified as acceptable or unacceptable. This feedback allows people to identify and correct errors in the use of the ECG device. The toolbox classifies the signals according to stationary, heart rate, and signal-to-noise ratio. If the limits of these three criteria are exceeded, this is indicated to the user. To develop and optimize the toolbox, two annotators reviewed a data set of 1,200 ECG leads to assess their quality, and each lead was classified as acceptable or unacceptable. The evaluation of the toolbox was done with a new data set of 4,200 leads, which were annotated the same way. This evaluation shows that the ECGAssess toolbox correctly classified over 94% of the 4,200 ECG leads as either acceptable or unacceptable in comparison to the annotations.

[Research on the detection algorithm of electrocardiogram characteristic wave based on energy segmentation and stationary wavelet transform]

The detection of electrocardiogram (ECG) characteristic wave is the basis of cardiovascular disease analysis and heart rate variability analysis. In order to solve the problems of low detection accuracy and poor real-time performance of ECG signal in the state of motion, this paper proposes a detection algorithm based on segmentation energy and stationary wavelet transform (SWT). Firstly, the energy of ECG signal is calculated by segmenting, and the energy candidate peak is obtained after moving average to detect QRS complex. Secondly, the QRS amplitude is set to zero and the fifth component of SWT is used to locate P wave and T wave. The experimental results show that compared with other algorithms, the algorithm in this paper has high accuracy in detecting QRS complex in different motion states. It only takes 0.22 s to detect QSR complex of a 30-minute ECG record, and the real-time performance is improved obviously. On the basis of QRS complex detection, the accuracy of P wave and T wave detection is higher than 95%. The results show that this method can improve the efficiency of ECG signal detection, and provide a new method for real-time ECG signal classification and cardiovascular disease diagnosis.

Combining Rhythm Information between Heartbeats and BiLSTM-Treg Algorithm for Intelligent Beat Classification of Arrhythmia

Arrhythmia is a cardiovascular disease that seriously affects human health. The identification and diagnosis of arrhythmia is an effective means of preventing most heart diseases. In this paper, a BiLSTM-Treg algorithm that integrates rhythm information is proposed to realize the automatic classification of arrhythmia. Firstly, the discrete wavelet transform is used to denoise the ECG signal, based on which we performed heartbeat segmentation and preserved the timing relationship between heartbeats. Then, different heartbeat segment lengths and the BiLSTM network model are used to conduct multiple experiments to select the optimal heartbeat segment length. Finally, the tree regularization method is used to optimize the BiLSTM network model to improve classification accuracy. And the interpretability of the neural network model is analyzed by analyzing the simulated decision tree generated in the tree regularization method. This method divides the heartbeat into five categories (nonectopic (N), supraventricular ectopic (S), ventricular ectopic (V), fused heartbeats (F), and unknown heartbeats (Q)) and is validated on the MIT-BIH arrhythmia database. The results show that the overall classification accuracy of the algorithm is 99.32%. Compared with other methods of classifying heartbeat, the BiLSTM-Treg network model algorithm proposed in this paper not only improves the classification accuracy and obtains higher sensitivity and positive predictive value but also has higher interpretability.

Heart rate variability and subsequent psychological distress among family members of intensive care unit patients

Objective

To determine whether heart rate variability (HRV; a physiological measure of acute stress) is associated with persistent psychological distress among family members of adult intensive care unit (ICU) patients.

Methods

This prospective study investigated family members of patients admitted to a study ICU. Participants’ variability in heart rate tracings were measured by low frequency (LF)/high frequency (HF) ratio and detrended fluctuation analysis (DFA). Questionnaires were completed 3 months after enrollment to ascertain outcome rates of anxiety, depression, and post-traumatic stress disorder (PTSD).

Results

Ninety-nine participants were enrolled (median LF/HF ratio, 0.92 [interquartile range, 0.64–1.38]). Of 92 participants who completed the 3-month follow-up, 29 (32%) had persistent anxiety. Logistic regression showed that LF/HF ratio (odds ratio [OR] 0.85, 95% confidence interval [CI] 0.43, 1.53) was not associated with 3-month outcomes. In an exploratory analysis, DFA α (OR 0.93, 95% CI 0.87, 0.99), α₁ (OR 0.97, 95% CI 0.94, 0.99), and α₂ (OR 0.94, 95% CI 0.88, 0.99) scaling components were associated with PTSD development.

Conclusion

Almost one-third of family members experienced anxiety at three months after enrollment. HRV, measured by LF/HF ratio, was not a predictor of psychologic distress, however, exploratory analyses indicated that DFA may be associated with PTSD outcomes.

A New Automatic QT-Interval Measurement Method for Wireless ECG Monitoring System Using Smartphone

QT-interval prolongation is an important parameter for heart arrhythmia diagnosis. It is the time interval from QRS-onset to the T-end of electrocardiogram (ECG). Manual measurement of QT-interval, especially for 12-leads ECG, is time-consuming. Hence, an automatic QT-interval measurement is necessary. A new method for automatic QT-interval measurement is presented in this paper, which mainly consists of three parts, including QRS-complex detection, determination of QRS-onset, and T-end determination. The QRS-complex detection is based on the modified Pan-Tompkins algorithm. The T-end is defined based on Region of Interest (ROI) maximum limit. We compare and test our proposed QT-interval measurement method with reference measurement in term of correlation coefficient and range of 95% LoA. The correlation coefficient and the range of 95% LoA are 0.575 and 0.290, respectively. The proposed method is successfully implemented in ECG monitoring system using smartphone with high performance. The accuracy, positive predictive, and sensitivity of the QRS-complex detection in the system are 99.70%, 99.78%, and 99.92%, respectively. The range of 95% LoA for the comparison between manual and the system’s QT-interval measurement is 0.216. The results show that the proposed method is dependable on the measure of the QT-interval and outperforms the other methods in term of correlation coefficient and range of 95% LoA.

ECG-based machine-learning algorithms for heartbeat classification

Electrocardiogram (ECG) signals represent the electrical activity of the human hearts and consist of several waveforms (P, QRS, and T). The duration and shape of each waveform and the distances between different peaks are used to diagnose heart diseases. In this work, to better analyze ECG signals, a new algorithm that exploits two-event related moving-averages (TERMA) and fractional-Fourier-transform (FrFT) algorithms is proposed. The TERMA algorithm specifies certain areas of interest to locate desired peak, while the FrFT rotates ECG signals in the time-frequency plane to manifest the locations of various peaks. The proposed algorithm’s performance outperforms state-of-the-art algorithms. Moreover, to automatically classify heart disease, estimated peaks, durations between different peaks, and other ECG signal features were used to train a machine-learning model. Most of the available studies uses the MIT-BIH database (only 48 patients). However, in this work, the recently reported Shaoxing People’s Hospital (SPH) database, which consists of more than 10,000 patients, was used to train the proposed machine-learning model, which is more realistic for classification. The cross-database training and testing with promising results is the uniqueness of our proposed machine-learning model.

Hidden Markov model-based heartbeat detector using electrocardiogram and arterial pressure signals

The automatic detection of a heartbeat is commonly performed by detecting the QRS complex in the electrocardiogram (ECG), however, various noise sources and missing data can jeopardize the reliability of the ECG. Therefore, there is a growing interest in combining the information from many physiological signals to accurately detect heartbeats. To this end, hidden Markov models (HMMs) are used in this work to jointly exploit the information from ECG, arterial blood pressure (ABP) and pulmonary arterial pressure (PAP) signals in order to conceive a heartbeat detector. After preprocessing the physiological signals, a sliding window is used to extract an observation sequence to be passed through two HMMs (previously trained on a training dataset) in order to obtain the log-likelihoods of observation and signals a detection if the difference of log-likelihoods exceeds an adaptive threshold. Several HMM-based heartbeat detectors were conceived to exploit the information from the ECG, ABP and PAP signals from the MIT-BIH Arrhythmia, PhysioNet Computing in Cardiology Challenge 2014, and MGH/MF Waveform databases. A grid search methodology was used to optimize the duration of the observation sequence and a multiplicative factor to form the adaptive threshold. Using the optimal parameters found on a training database through 10-fold cross-validation, sensitivity and positive predictivity above 99% were obtained on the MIT-BIH Arrhythmia and PhysioNet Computing in Cardiology Challenge 2014 databases, while they are above 95% in the MGH/MF waveform database using ECG and ABP signals. Our detector approach showed detection performances comparable with the literature in the three databases.

Supplementary Information

The online version contains supplementary material available at 10.1007/s13534-021-00192-x.

Machine learning model on heart rate variability metrics identifies asymptomatic toddlers exposed to zika virus during pregnancy

Objective. Although the Zika virus (ZIKV) seems to be prominently neurotropic, there are some reports of involvement of other organs, particularly the heart. Of special concern are those children exposed prenatally to ZIKV and born without microcephaly or other congenital anomalies. Electrocardiogram (ECG)-derived heart rate variability (HRV) metrics represent an attractive, low-cost, widely deployable tool for early identification of developmental functional alterations in exposed children born without such overt clinical symptoms. We hypothesized that HRV in such children would yield a biomarker of fetal ZIKV exposure. Our objective was to test this hypothesis in young children exposed to ZIKV during pregnancy.Approach. We investigated the HRV properties of 21 children aged 4-25 months from Brazil. The infants were divided into two groups, the ZIKV-exposed (n = 13) and controls (n = 8). Single-channel ECG was recorded in each child at ∼15 months of age and HRV was analyzed in 5 min segments to provide a comprehensive characterization of the degree of variability and complexity of the heart rate.Main results.Using a cubic support vector machine classifier we identified babies as Zika cases or controls with a negative predictive value of 92% and a positive predictive value of 86%. Our results show that a machine learning model derived from HRV metrics can help differentiate between ZIKV-affected, yet asymptomatic, and non-ZIKV-exposed babies. We identified the box count as the best HRV metric in this study allowing such differentiation, regardless of the presence of microcephaly.Significance.We show that it is feasible to measure HRV in infants and toddlers using a small non-invasive portable ECG device and that such an approach may uncover the memory ofin uteroexposure to ZIKV. We discuss putative mechanisms. This approach may be useful for future studies and low-cost screening tools involving this challenging to examine population.

PPGTempStitch: A MATLAB Toolbox for Augmenting Annotated Photoplethsmogram Signals

An annotated photoplethysmogram (PPG) is required when evaluating PPG algorithms that have been developed to detect the onset and systolic peaks of PPG waveforms. However, few publicly accessible PPG datasets exist in which the onset and systolic peaks of the waveforms are annotated. Therefore, this study developed a MATLAB toolbox that stitches predetermined annotated PPGs in a random manner to generate a long, annotated PPG signal. With this toolbox, any combination of four annotated PPG templates that represent regular, irregular, fast rhythm, and noisy PPG waveforms can be stitched together to generate a long, annotated PPG. Furthermore, this toolbox can simulate real-life PPG signals by introducing different noise levels and PPG waveforms. The toolbox can implement two stitching methods: one based on the systolic peak and the other on the onset. Additionally, cubic spline interpolation is used to smooth the waveform around the stitching point, and a skewness index is used as a signal quality index to select the final signal output based on the stitching method used. The developed toolbox is free and open-source software, and a graphical user interface is provided. The method of synthesizing by stitching introduced in this paper is a data augmentation strategy that can help researchers significantly increase the size and diversity of annotated PPG signals available for training and testing different feature extraction algorithms.

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.

Fetal Cardiovascular Decompensation During Labor Predicted From the Individual Heart Rate Tracing: A Machine Learning Approach in Near-Term Fetal Sheep Model

Background: When exposed to repetitive umbilical cord occlusions (UCO) with worsening acidemia, fetuses eventually develop cardiovascular decompensation manifesting as pathological hypotensive arterial blood pressure (ABP) responses to fetal heart rate (FHR) decelerations. Failure to maintain cardiac output during labor is a key event leading up to brain injury. We reported that the timing of the event when a fetus begins to exhibit this cardiovascular phenotype is highly individual and was impossible to predict. Objective: We hypothesized that this phenotype would be reflected in the individual behavior of heart rate variability (HRV) as measured by root mean square of successive differences of R-R intervals (RMSSD), a measure of vagal modulation of HRV, which is known to increase with worsening acidemia. This is clinically relevant because HRV can be computed in real-time intrapartum. Consequently, we aimed to predict the individual timing of the event when a hypotensive ABP pattern would emerge in a fetus from a series of continuous RMSSD data. Study Design: Fourteen near-term fetal sheep were chronically instrumented with vascular catheters to record fetal arterial blood pressure, umbilical cord occluder to mimic uterine contractions occurring during human labor and ECG electrodes to compute the ECG-derived HRV measure RMSSD. All animals were studied over a ~6 h period. After a 1-2 h baseline control period, the animals underwent mild, moderate, and severe series of repetitive UCO. We applied the recently developed machine learning algorithm to detect physiologically meaningful changes in RMSSD dynamics with worsening acidemia and hypotensive responses to FHR decelerations. To mimic clinical scenarios using an ultrasound-based 4 Hz FHR sampling rate, we recomputed RMSSD from FHR sampled at 4 Hz and compared the performance of our algorithm under both conditions (1,000 Hz vs. 4 Hz). Results: The RMSSD values were highly non-stationary, with four different regimes and three regime changes, corresponding to a baseline period followed by mild, moderate, and severe UCO series. Each time series was characterized by seemingly randomly occurring (in terms of timing of the individual onset) increase in RMSSD values at different time points during the moderate UCO series and at the start of the severe UCO series. This event manifested as an increasing trend in RMSSD values, which counter-intuitively emerged as a period of relative stationarity for the time series. Our algorithm identified these change points as the individual time points of cardiovascular decompensation with 92% sensitivity, 86% accuracy and 92% precision which corresponded to 14 ± 21 min before the visual identification. In the 4 Hz RMSSD time series, the algorithm detected the event with 3 times earlier detection times than at 1,000 Hz, i.e., producing false positive alarms with 50% sensitivity, 21% accuracy, and 27% precision. We identified the overestimation of baseline FHR variability by RMSSD at a 4 Hz sampling rate to be the cause of this phenomenon. Conclusions: The key finding is demonstration of FHR monitoring to detect fetal cardiovascular decompensation during labor. This validates the hypothesis that our HRV-based algorithm identifies individual time points of ABP responses to UCO with worsening acidemia by extracting change point information from the physiologically related fluctuations in the RMSSD signal. This performance depends on the acquisition accuracy of beat to beat fluctuations achieved in trans-abdominal ECG devices and fails at the sampling rate used clinically in ultrasound-based systems. This has implications for implementing such an approach in clinical practice.

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.

A Persistent Homology Approach to Heart Rate Variability Analysis With an Application to Sleep-Wake Classification

Persistent homology is a recently developed theory in the field of algebraic topology to study shapes of datasets. It is an effective data analysis tool that is robust to noise and has been widely applied. We demonstrate a general pipeline to apply persistent homology to study time series, particularly the instantaneous heart rate time series for the heart rate variability (HRV) analysis. The first step is capturing the shapes of time series from two different aspects—the persistent homologies and hence persistence diagrams of its sub-level set and Taken's lag map. Second, we propose a systematic and computationally efficient approach to summarize persistence diagrams, which we coined persistence statistics. To demonstrate our proposed method, we apply these tools to the HRV analysis and the sleep-wake, REM-NREM (rapid eyeball movement and non rapid eyeball movement) and sleep-REM-NREM classification problems. The proposed algorithm is evaluated on three different datasets via the cross-database validation scheme. The performance of our approach is better than the state-of-the-art algorithms, and the result is consistent throughout different datasets.

Heartbeat Classification Based on Multifeature Combination and Stacking-DWKNN Algorithm

Arrhythmia is one of the most common abnormal symptoms that can threaten human life. In order to distinguish arrhythmia more accurately, the classification strategy of the multifeature combination and Stacking-DWKNN algorithm is proposed in this paper. The method consists of four modules. In the preprocessing module, the signal is denoised and segmented. Then, multiple different features are extracted based on single heartbeat morphology, P length, QRS length, T length, PR interval, ST segment, QT interval, RR interval, R amplitude, and T amplitude. Subsequently, the features are combined and normalized, and the effect of different feature combinations on heartbeat classification is analyzed to select the optimal feature combination. Finally, the four types of normal and abnormal heartbeats were identified using the Stacking-DWKNN algorithm. This method is performed on the MIT-BIH arrhythmia database. The result shows a sensitivity of 89.42% and a positive predictive value of 94.90% of S-type beats and a sensitivity of 97.21% and a positive predictive value of 97.07% of V-type beats. The obtained average accuracy is 99.01%. Compared to other models with the same features, this method can improve accuracy and has a higher positive predictive value and sensitivity, which is important for clinical decision-making.

Platform for Analysis and Labeling of Medical Time Series

Reliable and diverse labeled reference data are essential for the development of high-quality processing algorithms for medical signals, such as electrocardiogram (ECG) and photoplethysmogram (PPG). Here, we present the Platform for Analysis and Labeling of Medical time Series (PALMS) designed in Python. Its graphical user interface (GUI) facilitates three main types of manual annotations-(1) fiducials, e.g., R-peaks of ECG; (2) events with an adjustable duration, e.g., arrhythmic episodes; and (3) signal quality, e.g., data parts corrupted by motion artifacts. All annotations can be attributed to the same signal simultaneously in an ergonomic and user-friendly manner. Configuration for different data and annotation types is straightforward and flexible in order to use a wide range of data sources and to address many different use cases. Above all, configuration of PALMS allows plugging-in existing algorithms to display outcomes of automated processing, such as automatic R-peak detection, and to manually correct them where needed. This enables fast annotation and can be used to further improve algorithms. The GUI is currently complemented by ECG and PPG algorithms that detect characteristic points with high accuracy. The ECG algorithm reached 99% on the MIT/BIH arrhythmia database. The PPG algorithm was validated on two public databases with an F1-score above 98%. The GUI and optional algorithms result in an advanced software tool that allows the creation of diverse reference sets for existing datasets.

Robust T-End Detection via T-End Signal Quality Index and Optimal Shrinkage

An automatic accurate T-wave end (T-end) annotation for the electrocardiogram (ECG) has several important clinical applications. While there have been several algorithms proposed, their performance is usually deteriorated when the signal is noisy. Therefore, we need new techniques to support the noise robustness in T-end detection. We propose a new algorithm based on the signal quality index (SQI) for T-end, coined as tSQI, and the optimal shrinkage (OS). For segments with low tSQI, the OS is applied to enhance the signal-to-noise ratio (SNR). We validated the proposed method using eleven short-term ECG recordings from QT database available at Physionet, as well as four 14-day ECG recordings which were visually annotated at a central ECG core laboratory. We evaluated the correlation between the real-world signal quality for T-end and tSQI, and the robustness of proposed algorithm to various additive noises of different types and SNR’s. The performance of proposed algorithm on arrhythmic signals was also illustrated on MITDB arrhythmic database. The labeled signal quality is well captured by tSQI, and the proposed OS denoising help stabilize existing T-end detection algorithms under noisy situations by making the mean of detection errors decrease. Even when applied to ECGs with arrhythmia, the proposed algorithm still performed well if proper metric is applied. We proposed a new T-end annotation algorithm. The efficiency and accuracy of our algorithm makes it a good fit for clinical applications and large ECG databases. This study is limited by the small size of annotated datasets.

Multichannel Esophageal Heart Rate Monitoring of Preterm Infants

OBJECTIVE

Autonomic dysregulation in preterm infants requires continuous monitoring of vital signs such as heart rate over days to months. Unfortunately, common surface electrodes are prone to electrocardiography (ECG) signal artifacts and cause serious skin irritations during long-term use. In contrast, esophageal ECG is known to be very sensitive due to the proximity of electrodes and heart and insensitive to external influences. This study addresses if multichannel esophageal ECG qualifies for heart rate monitoring in preterm infants.

METHODS

We recorded esophageal leads with a multi-electrode gastric feeding tube in a clinical study with 13 neonates and compared the heartbeat detection performance with standard surface leads. A computationally simple and versatile ECG wave detection algorithm was used.

RESULTS

Multichannel esophageal ECG manifested heartbeat sensitivity and positive predictive value greater than 98.5% and significant less false negative (FN) ECG waves as compared to surface ECG due to site-typical electrode motion artifacts. False positive bradycardia as indicated with more than 13 consecutive FN ECG waves was equally expectable in esophageal and surface channels. No adverse events were reported for the multi-electrode gastric feeding tube.

CONCLUSION

Heart rate monitoring of preterm infants with multiple esophageal electrodes is considered as feasible and reliable. Less signal artifacts will improve the detection of bradycardia, which is crucial for immediate interventions, and reduce alarm fatigue.

SIGNIFICANCE

Due to the possibility to integrate the multichannel ECG into a gastric feeding tube and meanwhile omit harmful skin electrodes, the presented system has great potential to facilitate future intensive care of preterm infants.

Exploratory data analysis based efficient QRS-complex detection technique with minimal computational load

Detection of QRS-complex in the electrocardiogram (ECG) plays a decisive role in cardiac disorder detection. We face many challenges in terms of powerline interference, baseline drift, and abnormal varying peaks. In this work, we propose an exploratory data analysis (EDA) based efficient QRS-complex detection technique with minimal computational load. This paper includes median and moving average filter for pre-processing of the ECG. The peak of filtered ECG is enhanced to third power of the signal. The root mean square (rms) of the signal is estimated for the decision making rule. This technique adapted the new concept for isoelectric line identification and EDA based QRS-complex detection. In this paper, total 10,70,981 beats were used for validation from MIT BIH-Arrhythmia Database (MIT-BIH), Fantasia Database (FDB), European ST-T database (ESTD), a self recorded dataset (SDB), and fetal ECG database (FTDB). Overall sensitivity of 99.65 % and positive predictivity rate of 99.84 % have been achieved. The proposed technique doesn't require selection, setting, and training for QRS-complex detection. Thus, this paper presents a QRS-complex detection technique based on simple decision rules.

Deep learning for comprehensive ECG annotation

BACKGROUND

Increasing utilization of long-term outpatient ambulatory electrocardiographic (ECG) monitoring continues to drive the need for improved ECG interpretation algorithms.

OBJECTIVE

The purpose of this study was to describe the BeatLogic® platform for ECG interpretation and to validate the platform using electrophysiologist-adjudicated real-world data and publicly available validation data.

METHODS

Deep learning models were trained to perform beat and rhythm detection/classification using ECGs collected with the Preventice BodyGuardian® Heart monitor. Training annotations were created by certified ECG technicians, and validation annotations were adjudicated by a team of board-certified electrophysiologists. Deep learning model classification results were used to generate contiguous annotation results, and performance was assessed in accordance with the EC57 standard.

RESULTS

On the real-world validation dataset, BeatLogic beat detection sensitivity and positive predictive value were 99.84% and 99.78%, respectively. Ventricular ectopic beat classification sensitivity and positive predictive value were 89.4% and 97.8%, respectively. Episode and duration F₁ scores (range 0–100) exceeded 70 for all 14 rhythms (including noise) that were evaluated. F₁ scores for 11 rhythms exceeded 80, 7 exceeded 90, and 5 including atrial fibrillation/flutter, ventricular tachycardia, ventricular bigeminy, ventricular trigeminy, and third-degree heart block exceeded 95.

CONCLUSION

The BeatLogic platform represents the next stage of advancement for algorithmic ECG interpretation. This comprehensive platform performs beat detection, beat classification, and rhythm detection/classification with greatly improved performance over the current state of the art, with comparable or improved performance over previously published algorithms that can accomplish only 1 of these 3 tasks.