ECGLight: lightweight on-device framework for paper ECG digitization and MI screening

Researchers have developed ECGLight, an end-to-end lightweight pipeline that converts smartphone photos of paper ECGs into calibrated 12-lead signals to screen for Myocardial Infarction (MI) pathologies. The system utilizes SHapley Additive exPlanations (SHAP) for interpretability and is designed to operate on devices with limited computational resources.

• Trained and evaluated on 21,799 ECGs from the PTB-XL dataset and validated on the hospital-acquired ECG-Matrix dataset.
• Runs in under 30 seconds per ECG on CPU-only resources.
• Achieves 95.51% accuracy (F1 = 0.9519) for MI detection on PTB-XL.
• Achieves 88.89% accuracy (F1 = 0.8862) for OMI detection on ECG-Matrix.

This work demonstrates that legacy paper records can be reliably digitized and analyzed in remote areas lacking digital export capabilities, high-speed connectivity, or high-end compute infrastructure.