A WGAN-GP-based ECG signal expansion and an SE-ResNet1D-based ECG classification method are proposed to address the problem of poor modeling results due to the imbalanced sample distribution of ECG data sets. The network architectures of WGAN-GP and SE-ResNet1D are designed according to the characteristics of ECG signals so that they can be better applied to the generation and classification of ECG signals. First, ECG data were generated using WGAN-GP on the MIT-BIH arrhythmia database to balance the dataset. Then, the experiments were performed using the AAMI category and inter-patient data partitioning principles, and classification experiments were performed using SE-ResNet1D on the imbalanced and balanced datasets, respectively, and compared with three networks, VGGNet, DenseNet and CNN+Bi-LSTM. The experimental results show that using WGAN-GP to balance the dataset can improve the accuracy and robustness of the model classification, and the proposed SE-ResNet1D outperforms the comparison model, with a precision of 95.80%, recall of 96.75% and an F1 measure of 96.27% on the balanced dataset. Our methods have the potential to be a useful diagnostic tool to assist cardiologists in the diagnosis of arrhythmias.

Background: Epicardial adipose tissue (EAT) essentially affects the pathophysiologic development of heart failure (HF), while existing research has not well elucidated its prognostic value for outcome. The present study aims at including relevant studies for systematically assessing its prognostic value in HF patients.Methods: The studies assessing the way EAT was employed to predict adverse HF outcomes were included from PubMed, Web of Science, Embase, and the Cochrane Library databases, and relevant data were accurately extracted. The primary outcome included the composite outcome of HF hospitalization and all-cause mortality (ACM). Secondary outcomes were the composite outcome of cardiovascular death and HF hospitalization and HF rehospitalization. We combined the standard mean difference (SMD) of EAT in HF patients with and without adverse events (AEs) and the EAT to adverse outcome hazard ratio (HR).Results: The nine included studies involved 1,939 HF patients and 329 control populations. HF patients with AEs presented a higher EAT vs. those without (SMD: 3.33, CI: 0.96–5.69, p = 0.006, I² = 98%). Increased EAT per unit indicated a higher risk of the composite outcome of HF hospitalization and all-cause mortality (HR: 1.28, CI: 1.42–1.85, p = 0.0002, I² = 93%) and HF readmission (HR: 1.05, CI: 1.03–1.07, p < 0.001, I² = 10%), but did not relate to that of cardiovascular death and HF hospitalization (HR: 1.17, CI: 0.99–1.39, p = 0.06, I² = 76%). The pooled AUC value for EAT to predict the primary outcome was 0.74 (CI: 0.66–0.82, p = 0.018, I² = 70.1%) in HF patients with EF >40%.Conclusion: EAT is considered a clinical predictor of the composite outcome of HF hospitalization and ACM and may contribute to the prediction of poor prognosis in HF patients.Systematic Review Registration: https://www.crd.york.ac.uk/, identifier [CRD420250653252].