ECG-Lens: Benchmarking ML & DL Models on PTB-XL Dataset

2026-04-21 · Source: cs.AI updates on arXiv.org · Field: Health & Wellbeing — Health & Medical Research, Medical Devices & Health Technology · Depth: Advanced, quick

Summary

The study "ECG-Lens: Benchmarking ML & DL Models on PTB-XL Dataset" compares three traditional machine learning algorithms (Decision Tree, Random Forest, Logistic Regression) and three deep learning models (Simple CNN, LSTM, Complex CNN/ECG-Lens) for automated classification of 12-lead electrocardiogram (ECG) signals. Utilizing the PTB-XL dataset, which includes recordings from normal and cardiac patients, the deep learning models were trained on raw ECG signals to automatically extract features. Data augmentation via Stationary Wavelet Transform (SWT) was applied to enhance performance and diversify training samples. Evaluation across metrics like accuracy, precision, recall, F1-score, and ROC-AUC revealed that the ECG-Lens model achieved the highest performance, with 80% classification accuracy and a 90% ROC-AUC, demonstrating the superior capability of complex CNNs over traditional methods for raw ECG data analysis.

Key takeaway

For AI Engineers developing cardiovascular diagnostic tools, this research indicates that deep learning architectures, particularly complex CNNs like ECG-Lens, offer superior performance on raw 12-lead ECG data compared to traditional machine learning. You should prioritize exploring and implementing advanced CNN models, potentially incorporating Stationary Wavelet Transform for data augmentation, to achieve higher accuracy and ROC-AUC in automated ECG classification systems.

Key insights

Deep learning models, especially complex CNNs, significantly outperform traditional ML for raw 12-lead ECG classification.

Principles

• Raw ECG data benefits from DL feature extraction.
• Data augmentation improves model generalization.

Method

The study trained ML/DL models on raw 12-lead ECG signals from the PTB-XL dataset, applying Stationary Wavelet Transform for data augmentation, and evaluated performance using multiple classification metrics.

In practice

• Use complex CNNs for ECG signal classification.
• Apply SWT for ECG data augmentation.

Topics

• ECG Signal Classification
• Deep Learning Models
• PTB-XL Dataset
• ECG-Lens Model
• Stationary Wavelet Transform

Best for: AI Engineer, AI Scientist, Machine Learning Engineer, Research Scientist

Related on AIssential

• ECG-Lens: Benchmarking ML & DL Models on PTB-XL Dataset — Complex CNNs significantly outperform traditional ML for automated 12-lead ECG classification on raw data.
• ECG Signal Classification Using Image-Based CNN Deep Learning — ECG signals can be effectively classified for cardiac abnormalities by transforming them into "2-D scalogram images" for CNN deep learning.
• A self-supervised electrocardiogram foundation model for empowering cardiovascular disease prediction and genetic factor discovery — A self-supervised ECG foundation model improves CVD prediction and uncovers novel genetic associations.
• A spectral audit framework reveals task-dependent aperiodic reliance across EEG and ECG deep learning — Deep learning models on physiological data often rely on aperiodic signals, impacting task performance.
• When is 3D Worth It? A Resource-Performance Frontier for CNNs and Transformers in Lung CT — Higher dimensionality in medical imaging models does not guarantee stable performance, especially with class imbalance.
• Python for Data Science & AI · Blog 18 of 20 — CNNs for Image Classification — Convolutional Neural Networks efficiently classify images by learning hierarchical features through local, reusable filters and weight sharing.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by cs.AI updates on arXiv.org.