Generalizable Beat-by-Beat Arrhythmia Detection by Using Weakly Supervised Deep Learning.

Beat-by-beat arrhythmia detection in ambulatory electrocardiogram (ECG) monitoring is critical for the evaluation and prognosis of cardiac arrhythmias, however, it is a highly professional demanding and time-consuming task. Current methods for automatic beat-by-beat arrhythmia detection suffer from poor generalization ability due to the lack of large-sample and finely-annotated (labels are given to each beat) ECG data for model training. In this work, we propose a weakly supervised deep learning framework for arrhythmia detection (WSDL-AD), which permits training a fine-grained (beat-by-beat) arrhythmia detector with the use of large amounts of coarsely annotated ECG data (labels are given to each recording) to improve the generalization ability. In this framework, heartbeat classification and recording classification are integrated into a deep neural network for end-to-end training with only recording labels. Several techniques, including knowledge-based features, masked aggregation, and supervised pre-training, are proposed to improve the accuracy and stability of the heartbeat classification under weak supervision. The developed WSDL-AD model is trained for the detection of ventricular ectopic beats (VEB) and supraventricular ectopic beats (SVEB) on five large-sample and coarsely-annotated datasets and the model performance is evaluated on three independent benchmarks according to the recommendations from the Association for the Advancement of Medical Instrumentation (AAMI). The experimental results show that our method improves the F 1 score of supraventricular ectopic beats detection by 8%-290% and the F1 of ventricular ectopic beats detection by 4%-11% on the benchmarks compared with the state-of-the-art methods of supervised learning. It demonstrates that the WSDL-AD framework can leverage the abundant coarsely-labeled data to achieve a better generalization ability than previous methods while retaining fine detection granularity. Therefore, this framework has a great potential to be used in clinical and telehealth applications. The source code is available at https://github.com/sdnjly/WSDL-AD.

A community effort to assess and improve computerized interpretation of 12-lead resting electrocardiogram.

Computerized interpretation of electrocardiogram plays an important role in daily cardiovascular healthcare. However, inaccurate interpretations lead to misdiagnoses and delay proper treatments. In this work, we built a high-quality Chinese 12-lead resting electrocardiogram dataset with 15,357 records, and called for a community effort to improve the performances of CIE through the China ECG AI Contest 2019. This dataset covers most types of ECG interpretations, including the normal type, 8 common abnormal types, and the other type which includes both uncommon abnormal and noise signals. Based on the Contest, we systematically assessed and analyzed a set of top-performing methods, most of which are deep neural networks, with both their commonalities and characteristics. This study establishes the benchmarks for computerized interpretation of 12-lead resting electrocardiogram and provides insights for the development of new methods. Graphical Abstract A community effort to assess and improve computerized interpretation of 12-lead resting electrocardiogram.

Automatic Multi-Label ECG Classification with Category Imbalance and Cost-Sensitive Thresholding.

Automatic electrocardiogram (ECG) classification is a promising technology for the early screening and follow-up management of cardiovascular diseases. It is, by nature, a multi-label classification task owing to the coexistence of different kinds of diseases, and is challenging due to the large number of possible label combinations and the imbalance among categories. Furthermore, the task of multi-label ECG classification is cost-sensitive, a fact that has usually been ignored in previous studies on the development of the model. To address these problems, in this work, we propose a novel deep learning model-based learning framework and a thresholding method, namely category imbalance and cost-sensitive thresholding (CICST), to incorporate prior knowledge about classification costs and the characteristic of category imbalance in designing a multi-label ECG classifier. The learning framework combines a residual convolutional network with a class-wise attention mechanism. We evaluate our method with a cost-sensitive metric on multiple realistic datasets. The results show that CICST achieved a cost-sensitive metric score of 0.641 ± 0.009 in a 5-fold cross-validation, outperforming other commonly used thresholding methods, including rank-based thresholding, proportion-based thresholding, and fixed thresholding. This demonstrates that, by taking into account the category imbalance and predefined cost information, our approach is effective in improving the performance and practicability of multi-label ECG classification models.

Effects of long-term fasting and confinement on the cardiovascular activity.

Fasting has been demonstrated to improve health and slow aging in human and other species; however, its impact on the human body in the confined environment is still unclear. This work studies the effects of long-term fasting and confined environment on the cardiovascular activities of human via a 10-day fasting experiment with two groups of subjects being in confined (6 subjects) and unconfined (7 subjects) environments respectively and undergoing the same four-stage fasting/feeding process. It is found that the confinement has significant influences on the autonomic regulation to the heart rate during the fasting process by altering the activity of the parasympathetic nervous system, which is manifested by the significant higher pNN50, rMSSD, and Ln-HF of heart rate variability (HRV) (p < 0.05) and slower heart rate (p < 0.01) in the confined group than that in the unconfined group. Furthermore, the long-term fasting induces a series of changes in both groups, including reduced level of serum sodium (p < 0.01), increased the serum calcium (p < 0.05), prolonged QTc intervals (p < 0.05), and reduced systolic blood pressures (p < 0.05). These effects are potentially negative to human health and therefore need to be treated with caution. Study of the effects of fasting and confinement on the cardiovascular activities.

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.

Three-Dimensional Liver Image Segmentation Using Generative Adversarial Networks Based on Feature Restoration.

Medical imaging provides a powerful tool for medical diagnosis. In the process of computer-aided diagnosis and treatment of liver cancer based on medical imaging, accurate segmentation of liver region from abdominal CT images is an important step. However, due to defects of liver tissue and limitations of CT imaging procession, the gray level of liver region in CT image is heterogeneous, and the boundary between the liver and those of adjacent tissues and organs is blurred, which makes the liver segmentation an extremely difficult task. In this study, aiming at solving the problem of low segmentation accuracy of the original 3D U-Net network, an improved network based on the three-dimensional (3D) U-Net, is proposed. Moreover, in order to solve the problem of insufficient training data caused by the difficulty of acquiring labeled 3D data, an improved 3D U-Net network is embedded into the framework of generative adversarial networks (GAN), which establishes a semi-supervised 3D liver segmentation optimization algorithm. Finally, considering the problem of poor quality of 3D abdominal fake images generated by utilizing random noise as input, deep convolutional neural networks (DCNN) based on feature restoration method is designed to generate more realistic fake images. By testing the proposed algorithm on the LiTS-2017 and KiTS19 dataset, experimental results show that the proposed semi-supervised 3D liver segmentation method can greatly improve the segmentation performance of liver, with a Dice score of 0.9424 outperforming other methods.

Heart failure-induced atrial remodelling promotes electrical and conduction alternans.

Heart failure (HF) is associated with an increased propensity for atrial fibrillation (AF), causing higher mortality than AF or HF alone. It is hypothesized that HF-induced remodelling of atrial cellular and tissue properties promotes the genesis of atrial action potential (AP) alternans and conduction alternans that perpetuate AF. However, the mechanism underlying the increased susceptibility to atrial alternans in HF remains incompletely elucidated. In this study, we investigated the effects of how HF-induced atrial cellular electrophysiological (with prolonged AP duration) and tissue structural (reduced cell-to-cell coupling caused by atrial fibrosis) remodelling can have an effect on the generation of atrial AP alternans and their conduction at the cellular and one-dimensional (1D) tissue levels. Simulation results showed that HF-induced atrial electrical remodelling prolonged AP duration, which was accompanied by an increased sarcoplasmic reticulum (SR) Ca2+ content and Ca2+ transient amplitude. Further analysis demonstrated that HF-induced atrial electrical remodelling increased susceptibility to atrial alternans mainly due to the increased sarcoplasmic reticulum Ca2+-ATPase (SERCA) Ca2+ reuptake, modulated by increased phospholamban (PLB) phosphorylation, and the decreased transient outward K+ current (Ito). The underlying mechanism has been suggested that the increased SR Ca2+ content and prolonged AP did not fully recover to their previous levels at the end of diastole, resulting in a smaller SR Ca2+ release and AP in the next beat. These produced Ca2+ transient alternans and AP alternans, and further caused AP alternans and Ca2+ transient alternans through Ca2+→AP coupling and AP→Ca2+ coupling, respectively. Simulation of a 1D tissue model showed that the combined action of HF-induced ion channel remodelling and a decrease in cell-to-cell coupling due to fibrosis increased the heart tissue's susceptibility to the formation of spatially discordant alternans, resulting in an increased functional AP propagation dispersion, which is pro-arrhythmic. These findings provide insights into how HF promotes atrial arrhythmia in association with atrial alternans.

Automatic Detection of Atrial Fibrillation Based on Continuous Wavelet Transform and 2D Convolutional Neural Networks.

Atrial fibrillation (AF) is the most common cardiac arrhythmias causing morbidity and mortality. AF may appear as episodes of very short (i.e., proximal AF) or sustained duration (i.e., persistent AF), either form of which causes irregular ventricular excitations that affect the global function of the heart. It is an unmet challenge for early and automatic detection of AF, limiting efficient treatment strategies for AF. In this study, we developed a new method based on continuous wavelet transform and 2D convolutional neural networks (CNNs) to detect AF episodes. The proposed method analyzed the time-frequency features of the electrocardiogram (ECG), thus being different to conventional AF detecting methods that implement isolating atrial or ventricular activities. Then a 2D CNN was trained to improve AF detection performance. The MIT-BIH Atrial Fibrillation Database was used for evaluating the algorithm. The efficacy of the proposed method was compared with those of some existing methods, most of which implemented the same dataset. The newly developed algorithm using CNNs achieved 99.41, 98.91, 99.39, and 99.23% for the sensitivity, specificity, positive predictive value, and overall accuracy (ACC) respectively. As the proposed algorithm targets the time-frequency feature of ECG signals rather than isolated atrial or ventricular activity, it has the ability to detect AF episodes for using just five beats, suggesting practical applications in the future.