Search

Usually, manual labeling of large-scale electrocardiograms (ECGs) for deep learning is always expensive, as it requires considerable effort and time from cardiologists. Currently, contrastive learning can utilize unlabeled ECGs to improve deep learning using insufficient labels. However, this field may still lack instructive literature. In this paper, a dense lead contrast (DLC) is proposed for effective contrastive learning on multilead ECGs. It develops contrastive learning between any two leads from different views to explore intralead and interlead invariance. A joint loss combining intralead and interlead contrastive loss guides the DLC pretraining. Moreover, DLC introduces a multibranch network (MBN) for contrastive learning, generates a representation for each lead, and fuses all the leads for a global representation. In the downstream tasks, DLC outperforms the standard contrastive learning paradigm of multilead ECGs by 2.78%∼6.59% in AUROC (linear probe). Using only 10% of the labeled training data, it still outperforms standard contrastive learning by a significant margin in AUROC. Compared with existing methods, DLC shows obvious advantages in all the experiments. Therefore, DLC may be more suitable for the contrastive learning of multilead ECGs. Its good performance based on insufficient labels can alleviate the cardiologists' burden from data labeling. • Dense lead contrast (DLC) conducts an effective self-supervised pre-training. • DLC jointly explores intra- and inter-lead invariance of electrocardiograms (ECGs). • Multiple branch network (MBN) architecture organizes all the leads. • Raw ECGs from a real-world hospital are used to validate algorithm robustness. [ABSTRACT FROM AUTHOR]

Graph-based models for Electrocardiogram (ECG) incorporate physiological spatial information among ECG leads in elegant manners. It is unachievable for convolution-based models. Nevertheless, existing ECG graph models heavily rely on manually pre-designed structures. Such structures may not be optimal for all ECGs. Moreover, these models lag behind larger representative models, despite their superiority in lightweight cases. In this study, a novel graph structure learning method is proposed. It can learn optimal structures and enhance model representations simultaneously. Typically, ECGs belong to limb/chest lead systems. To explore inter-system and intra-system connections, heuristic optimization including genetic algorithm and particle swarm optimization is employed. Specifically, candidate nodes/edges are encoded as individuals. Thereby graph structure learning is transformed into population iterations. In this scenario, graph-level embedding becomes non-applicable because models require retraining once structures change. Instead, node-level embedding continuously enhances models during structure learning. Consequently, Honest-GE further exploits graph methods' superiority in lightweight cases. It outperforms state-of-the-art models 11.53% and 10.03% in F 1 on two databases. Additionally, it demonstrates comparable performance with larger models. In general, graph learning results corroborate medical knowledge and offer insights for lead selection. Honest-GE provides a promising avenue for high-performing lightweight models and portable deployments with fewer leads. • Heuristic algorithms are applied for Electrocardiogram graph structure learning. • Node-embedding GraphSAGE is employed to enhance model continuously in searching. • Honest-GE further shows superiority over existing methods with similar parameters. • Honest-GE achieves competent performance against representative large models. • Result corroborate with medical knowledge and provide advice for lead selection. [ABSTRACT FROM AUTHOR]

Deep cardiovascular disease (CVD) detection usually achieves good performance with large-scale labeled electrocardiograms (ECGs), but manual labeling of ECGs is tedious. Semi-supervised learning (SSL) aims to improve model performance through unlabeled data. In this study, an adaptive threshold-based semi-supervised learning model (ATSS-LGP) is proposed. It introduces the multibranch network (MBN) to generate local and global predictions for 12-lead ECG. The labeled ECGs are used to train the model in supervised manner and produce adaptive thresholds through the proposed prediction voting decision mechanism. The unlabeled ECGs are divided into high-confidence and low-confidence parts by the adaptive thresholds. The pseudo-labeling technique and consistency regularization are used to jointly guide the unsupervised learning, which can fully utilize all unlabeled ECGs. To the best of our knowledge, this is the first SSL algorithm that explores adaptive threshold in ECG. ATSS-LGP shows impressive performance in CVD detection. Using the same number of labeled ECGs, it achieves at least a 5.15% increase in accuracy over pure supervised learning. Moreover, ATSS-LGP achieves comparable performance to fully supervised method using only 10% of labeled ECGs. In summary, ATSS-LGP is a suitable SSL algorithm for 12-lead ECGs, which can greatly reduce the burden of ECG labeling in deep learning. [ABSTRACT FROM AUTHOR]

• Shockable and non-shockable ECG signals are analyzed. • A review study for shockable ECG signal recognition is discussed. • Unique bispectrum and recurrence plots are proposed for shockable and non-shockable ECG signals. • Performance of nonlinear features for differentiating shockable and non-shockable. • ECG signals is compared. Sudden cardiac death from lethal arrhythmia is a preventable cause of death. Ventricular fibrillation and tachycardia are shockable electrocardiographic (ECG)rhythms that can respond to emergency electrical shock therapy and revert to normal sinus rhythm if diagnosed early upon cardiac arrest with the restoration of adequate cardiac pump function. However, manual inspection of ECG signals is a difficult task in the acute setting. Thus, computer-aided arrhythmia classification (CAAC) systems have been developed to detect shockable ECG rhythm. Traditional machine learning and deep learning methods are now progressively employed to enhance the diagnostic accuracy of CAAC systems. This paper reviews the state-of-the-art machine and deep learning based CAAC expert systems for shockable ECG signal recognition, discussing their strengths, advantages, and drawbacks. Moreover, unique bispectrum and recurrence plots are proposed to represent shockable and non-shockable ECG signals. Deep learning methods are usually more robust and accurate than standard machine learning methods but require big data of good quality for training. We recommend collecting large accessible ECG datasets with a meaningful proportion of abnormal cases for research and development of superior CAAC systems. [ABSTRACT FROM AUTHOR]

This paper presents a reconfigurable front-end (FE) circuit for acquiring various low-frequency biomedical signals. An energy and area-efficient tunable filter is proposed for adapting the FE bandwidth to the signal of interest. The filter is designed using a switched-R-MOSFET-C (SRMC) technique to realize the needed ultra-low cutoff frequency. An 8-bit SAR ADC, following the filter, quantizes the signal, while the SAR control logic is re-used to accurately program the filter bandwidth from 40 Hz to 320 Hz with a 40 Hz step. The prototype chip includes the complete FE system, formed of an instrumentation amplifier (IA), a programmable-gain amplifier (PGA), and the proposed tunable filter followed by the SAR ADC. Implemented in 0.13 $\mu \text{m}$ CMOS technology, the IC occupies a 0.6 mm2 active area while consuming 6.3 $\mu \text{W}$ dc power from a 2-V supply. Measurement results show a FE gain range of 43–55 dB with an integrated input-referred noise (${V_{\text {IRN}}}$) of 3.45 $\mu V_{\text {rms}}$ , a 66 dB dynamic range (DR), and a total-harmonic distortion (THD) of −68 dB at an input amplitude of 6 $\text{m}V_{PP}$. The effective number of bits (ENOB) for the ADC is 7.921 bits at 1-kS/s. In real-time Electrocardiogram (ECG), Electromyography (EMG), and Electroencephalography (EEG) measurements, high-fidelity waveforms are acquired using the proposed FE IC, validating the system’s reconfigurability and high-linearity. [ABSTRACT FROM AUTHOR]

• A novel ECG Diagnosis model for mobile devices is proposed based on dynamic neural network. • The model has high inference efficiency and device versatility opposed to static models. • Two multi-classifier collaborative learning methods are proposed. • A sigmoid-based early exit criterion is designed for multi-label ECG diagnosis. Mobile health devices with automatic electrocardiogram diagnosis models facilitate long-term cardiac monitoring and enhance the sensitivity of detecting paroxysmal cardiovascular disease. However, these devices often encounter limitations in terms of computational power and battery life. Moreover, iterations of mobile health device versions may result in changes to their computational limitations, leading to additional costs for model design and training. Therefore, it is crucial to develop lightweight models that enable efficient inference under resource constraints and are compatible with various devices. To address these challenges, we propose a resource-efficient and device-adaptive electrocardiogram diagnosis model based on a multi-classifier dynamic neural network. This model achieves efficient computation by exiting simple samples early and can adapt to the computational limits of various devices by adjusting predefined thresholds. Furthermore, we propose two novel approaches, namely rethinking structure and feature-based knowledge distillation, which enhance the collaboration among classifiers and improve the overall performance. The proposed model is evaluated using three electrocardiogram datasets: PTB-XL (AUC = 0.9706), ICBEB2018 (AUC = 0.964), and MIT-BIH arrhythmia (ACC = 0.994, F1 = 0.961). The results consistently demonstrate the superiority of the proposed model over the comparative models across all three datasets, with average improvements of 2.4%, 0.5%, and 1.3% respectively. [ABSTRACT FROM AUTHOR]

The electrocardiogram (ECG) signal is made up of sequences of three distinct waves, including the P-wave, QRS-complex, and T-wave. These sequences may contain several different varieties of feature representation, which means various appearances represent different attribute knowledge of heart activity. Consequently, delicate analysis of the ECG attribute knowledge is critical, which focuses on empowering computers to understand ECG signals in a way physicians do. Traditional ECG systems usually only focus on the key wave detection or sub-sequences morphological classification, and the complex ECG features make the analysis process difficult to handle both tasks. In this paper, our goal is to propose a novel ECG analysis technique for ECG medical attribute knowledge feature extraction, termed ECG-MAKE method. The ECG-MAKE method mainly addresses two challenging issues—ECG signal key points detection and morphological delineation—by introducing four modules: signal preprocessing, key point detection, adaptive error correction, and attribute knowledge extraction. Signal preprocessing is based on a wavelet multi-resolution algorithm, which extracts both time and frequency domain information simultaneously, and achieves the goal of enhancing raw signals by filtering signals below 1 Hz and above 100 Hz. The key point detection combines mean-value difference and sliding window to detect R-peak. And according to the R-peak, further utilizes the ECG rules and threshold to detect the onset, peak, and offset of different heartbeat waveforms such as the P-waves and T-waves. Then, for standard 12-lead ECG, the adaptive error correction can automatically correct the spatial position of certain leads missed and false detection key points. Finally, the attribute knowledge extraction module is based on the statistical features of key points to obtain the ECG attributes knowledge of clinical diagnosis of ECG. Extensive experiments conducted on the publicly available database QT database (99.57% F1-score) have demonstrated the effectiveness of the proposed ECG-MAKE method. The proposed method has a higher classification performance than current state-of-the-art methods with an F1-score performance of 94.30% on the ZZU-ECG database. • A medical attribute knowledge extraction method is proposed to delineate ECG signal. • A strategy is proposed to correct missed and false detections of ECG knowledge. • A framework is proposed to provide ECG refined annotation for deep learning model. • High classification performance compared to current state-of-the-art methods. [ABSTRACT FROM AUTHOR]

Classification of electrocardiogram (ECG) data stream is essential to diagnosis of critical heart conditions. It is vital to accurately detect abnormality in the ECG in order to prevent possible beginning of life-threatening cardiac symptoms. In this paper, we focus on identifying premature ventricular contraction (PVC) which is one of the most common heart rhythm abnormalities. We use “Replacing” strategy to check the effects of each individual heartbeat on the variation of principal directions. Based on this idea, an online PVC detection method is proposed to classify the new arriving PVC beats in the real-time and online manner. The proposed approach is tested on the MIT-BIH arrhythmia database (MIT-BIH-AR). The PVC detection accuracy was 98.77%, with the sensitivity and positive predictivity of 96.12% and 86.48%, respectively. These results are an improvement on previous reported results for PVC detection. In addition, our proposed method is effective in terms of computation time. The average execution time of our proposed method was 3.83 s for a 30 min ECG recording. It shows the capability of the classifier to detect abnormal PVCs in online manner. [ABSTRACT FROM AUTHOR]

Abstract: This paper presents an essentially non-oscillatory point-value (ENOPV) scheme based ECG compression. ENOPV scheme is a combination of multiresolution scheme (analysis) and interpolation scheme (synthesis). Advantage of using ENOPV scheme is that it avoids the discontinuities and therefore large numbers of coefficient do not appear at edges. It also provides better compression capabilities. The results are presented in tabular and graphical form for the different ECG signals of varying characteristics. Results are also compared with the results reported in the literature. It has been observed that there is significant increase in compression ratio (CR) at a given quality. For example, at percentage of root mean square difference (PRD) 2.65, the CR in case of ENOPV based scheme is 28.22, which is higher than 10.84, the best reported result. [Copyright &y& Elsevier]

Abstract: Heart rate variability (HRV), a widely adopted quantitative marker of the autonomic nervous system can be used as a predictor of risk of cardiovascular diseases. Moreover, decreased heart rate variability (HRV) has been associated with an increased risk of cardiovascular diseases. Hence in this work HRV signal is used as the base signal for predicting the risk of cardiovascular diseases. The present study concerns nine cardiac classes that include normal sinus rhythm (NSR), congestive heart failure (CHF), atrial fibrillation (AF), ventricular fibrillation (VF), preventricular contraction (PVC), left bundle branch block (LBBB), complete heart block (CHB), ischemic/dilated cardiomyopathy (ISCH) and sick sinus syndrome (SSS). A total of 352 cardiac subjects belonging to the nine classes were analyzed in the frequency domain. The fast Fourier transforms (FFT) and three other modeling techniques namely, autoregressive (AR) model, moving average (MA) model and the autoregressive moving average (ARMA) model are used to estimate the power spectral densities of the RR interval variability. The spectral parameters obtained from the spectral analysis of the HRV signals are used as the input parameters to the artificial neural network (ANN) for classification of the different cardiac classes. Our findings reveal that the ARMA modeling technique seems to give better resolution and would be more promising for clinical diagnosis. [Copyright &y& Elsevier]

In this paper, we use the multidimensional multiscale parser (MMP) algorithm, a recently developed universal lossy compression method, to compress data from electrocardiogram (ECG) signals. The MMP is based on approximate multiscale pattern matching, encoding segments of an input signal using expanded and contracted versions of patterns stored in a dictionary. The dictionary is updated using concatenated and displaced versions of previously encoded segments, therefore MMP builds its own dictionary while the input data is being encoded. The MMP can be easily adapted to compress signals of any number of dimensions, and has been successfully applied to compress two-dimensional (2-D) image data. The quasi-periodic nature of ECG signals makes them suitable for compression using recurrent patterns, like MMP does. However, in order for MMP to be able to efficiently compress ECG signals, several adaptations had to be performed, such as the use of a continuity criterion among segments and the adoption of a prune-join strategy for segmentation. The rate-distortion performance achieved was very good. We show simulation results were MMP performs as well as some of the best encoders in the literature, although at the expense of a high computational complexity. [ABSTRACT FROM AUTHOR]

A new digital programmable CMOS analog front-end (AFE) IC for measuring electroencephalograph or electrocardiogram signals in a portable instrumentation design approach is presented. This includes a new high-performance rail-to-rail instrumentation amplifier (IA) dedicated to the low-power AFE IC. The measurement results have shown that the proposed biomedical AFE IC, with a die size of 4.81 mm2, achieves a maximum stable ac gain of 10000 V/V input-referred noise of 0.86 μ Vrms (0.3 Hz-150 Hz), common-mode rejection ratio of at least 115 dB (0-1 kHz), input-referred de offset of less than 60 μV, input common mode range from -1.5 V to 1.3 V, and current drain of 485 μA (excluding the power dissipation of external clock oscillator) at a ±1.5-V supply using a standard 0.5-μm CMOS process technology. [ABSTRACT FROM AUTHOR]

• We conceive the similarity in the phase space of chaotic dynamics. • We derive a fast-convergent algorithm for extracting ECG signals based on the proposed similarity index. • We numerically demonstrate the simplicity and effectiveness of the proposed scheme for the blind extraction of ECG signals. Electrocardiogram (ECG), as a biological signal that contains important information about the cardiac activities of heart, exhibits chaotic characteristics. Since a clean ECG signal is of vital importance in the diagnosis and analysis of heart diseases, we address the task of extracting ECG for a set of noisy observations. Based on the phase space similarity of ECG, we propose an objective called similarity index which fully describes this similarity. A low-complexity algorithm is presented for the similarity index. Simulation results confirm the effectiveness of the proposed method by making comparisons with conventional benchmarks. [ABSTRACT FROM AUTHOR]