Diffusion-Based ECG Augmentation
Generate synthetic ECG signals via style transfer between Atrial Fibrillation (AFib) and Normal Sinus Rhythm classes using a disentangled diffusion model.
How it works: The model separates ECG signals into content (beat morphology) and style (rhythm), then generates new ECGs by transferring the style of one class onto another using SDEdit denoising.
Key finding: Replacing 33% of real training data with these synthetic ECGs produces statistically equivalent classifier performance (TOST, p = 0.007).
Select a pre-loaded ECG to see the augmentation result:
Browse Preprocessed ECG Data
Explore 200 curated ECGs from the MIMIC-IV test set (100 Normal + 100 AFib).
Select an ECG to view its signal and metadata, then download it as a .npy file to use in the Upload ECG tab.
Click Load to view available ECGs.
ECG Table
Select an ECG from the table to see its details.