ECG-Lens: Benchmarking ML & DL Models on PTB-XL Dataset
Summary: arXiv:2604.15822v1 Announce Type: cross
Abstract
Automated classification of electrocardiogram (ECG) signals is a useful tool for diagnosing and monitoring cardiovascular diseases. This study compares three traditional machine learning algorithms (Decision Tree Classifier, Random Forest Classifier, and Logistic Regression) and three deep learning models (Simple Convolutional Neural Network (CNN), Long Short-Term Memory (LSTM), and Complex CNN (ECGLens)) for the classification of ECG signals from the PTB-XL dataset, which contains 12-lead recordings from normal patients and patients with various cardiac conditions.
The deep learning models were trained on raw ECG signals, allowing them to automatically extract discriminative features. Data augmentation using the Stationary Wavelet Transform (SWT) was applied to enhance model performance, increase the diversity of training samples, and preserve the essential characteristics of the ECG signals. The models were evaluated using multiple metrics, including accuracy, precision, recall, F1-score, and ROC-AUC.
Key Findings
- The ECG-Lens model achieved the highest performance with:
- 80% classification accuracy
- 90% ROC-AUC
- Deep learning architectures, particularly complex CNNs, substantially outperform traditional ML methods on raw 12-lead ECG data.
- The study provides a practical benchmark for selecting automated ECG classification models.
- Identifies directions for condition-specific model development.
Conclusion
The findings underscore the potential of deep learning models in transforming the landscape of ECG analysis and cardiovascular disease diagnosis. By leveraging the extensive capabilities of complex CNN architectures, healthcare professionals can improve diagnostic accuracy and patient outcomes. This research not only showcases the efficacy of advanced machine learning techniques but also sets the stage for future explorations in automated cardiac signal classification.
Future Directions
As the field continues to evolve, several key areas warrant further investigation:
- Integration of additional data sources for enhanced model training.
- Exploration of hybrid models that combine traditional and deep learning approaches.
- Real-time application of these models in clinical settings.
- Investigation of interpretability in deep learning models for better clinician understanding.
This research lays a foundation for future advancements in automated ECG classification, paving the way for the development of more robust and effective diagnostic tools in cardiology.