Skip to content

Dataset · edcci

PTB-XL-based electrocardiogram image dataset

edcci/GenECG

Synthetic ECG images (clean and corrupted) for computer vision training.

What's inside

This dataset converts ECG signals from the well-known PTB-XL medical database (21,799 unique records) into image format. The collection is divided into subsets: for example, Subset A contains 21,799 clean ECG images without visual distortions, while Subset B includes images with defects. Published by edcci specifically for computer vision tasks involving cardiogram analysis.

What you can do with it

  • Training computer vision models for diagnosis using ECG scans and photos
  • Testing the robustness of medical classifiers against visual noise and defects
  • Comparing the effectiveness of ECG analysis based on signals versus ready images
Downloads this month
36.3K
Likes
4

How to load it

from datasets import load_dataset

ds = load_dataset("edcci/GenECG")