Key result
TG-MS-GAN framework achieves up to ~99% accuracy in ECG arrhythmia detection by addressing class imbalance.
Why the study?
ECG-based arrhythmia detection is challenged by severe class imbalance, morphological variability, and wearable noise, with traditional augmentation distorting waveforms and classifiers underperforming on minority categories.
A novel morphology-preserving generative adversarial network framework demonstrates high accuracy and robustness for ECG arrhythmia detection across multiple heterogeneous datasets, effectively addressing class imbalance.
Answers are grounded in this paper and cite their sources.
May aid arrhythmia detection amid class imbalance; leaves open prospective clinical validation before practice change.
Electrocardiogram (ECG) based arrhythmia detection remains challenging due to severe class imbalance, morphological variability, and noise commonly present in wearable monitoring environments. Traditional augmentation techniques often distort clinically meaningful waveform characteristics, while deep learning classifiers typically underperform on minority arrhythmia categories. This study proposes a unified morphology-preserving framework that integrates a Transfer-Guided Multi-Stage Generative Adversarial Network (TG-MS-GAN), an EfficientNetV2-S classifier, and an adaptive discriminative loss to enhance minority-class representation and improve overall diagnostic robustness. The TG-MS-GAN progressively reconstructs ECG beats across multi-resolution stages, synthesizing high-fidelity minority-class waveforms that preserve P-wave integrity, QRS morphology, and ST-segment structure. A transfer-guidance mechanism introduces physiological priors into early-stage generators, stabilizing convergence under extreme sample scarcity. EfficientNetV2-S serves as a lightweight yet expressive backbone capable of capturing fine-grained temporal features with low computational cost, making the method suitable for real-time wearable deployment. An adaptive hybrid loss combining weighted cross-entropy, center loss, and difficulty-based focusing improves inter-class separability and stabilizes optimization under severe imbalance. Experimental results demonstrate consistent performance across different ECG classification tasks, achieving accuracies of 99.1%, 97.6%, 91.4%, and 99.0% on the MIT-BIH, Chapman-Shaoxing, PTB-XL, and Fetal ECG datasets, respectively. The corresponding Macro-F1 scores reach 0.987, 0.962, 0.839, and 0.989, while the AUPRC values attain 0.989, 0.891, 0.846, and 0.991, respectively. These results indicate that the proposed framework effectively addresses class imbalance while maintaining robust classification performance across heterogeneous ECG datasets. The source code is available at https://github.com/phdmajidsepahvand/TG-MS-GAN-ECG.
No takes yet. Share an insight, caveat, or question.
Sepahvand et al. (2026) studied ECG arrhythmia. TG-MS-GAN framework with EfficientNetV2-S was evaluated on ECG classification performance (Accuracy, Macro-F1, AUPRC). The TG-MS-GAN framework achieved classification accuracies of 91.4% to 99.1% across four heterogeneous ECG datasets, effectively addressing class imbalance in arrhythmia detection.
Synapse has enriched 10 related papers. Consider them for comparative context: