University of Oxford (Krones et al.)
Reconstructs digital 12-lead ECG waveforms from scanned or photographed paper printouts, using an nnU-Net image segmentation model to trace the signal pixels followed by a Hough-transform-based reconstruction pipeline. This is a digitization tool rather than a diagnostic model - it recovers a usable signal from a paper record rather than producing a diagnosis. Won the PhysioNet/Computing in Cardiology Challenge 2024; developed by a team at the University of Oxford.
Architecture
CNN (2D)
nnU-Net image segmentation of ECG trace pixels from a scanned/printed page image, combined with a Hough-transform-based lead/gridline detection and signal-reconstruction pipeline
Framework
PyTorch
Added to catalog
2026-07-10
BSD 2-Clause
License for model weights only. Associated code may be licensed seperately, check code source for specific terms.
52% male / 48% female; age range 0-95 (median ~62). 21,837 10-second 12-lead ECG records.
Digitized 12-lead ECG voltage-time signal reconstructed from a scanned/photographed paper printout (internally via nnU-Net signal-trace segmentation followed by Hough-transform reconstruction)