README.md

---
configs:
- config_name: default
  data_files:
  - path: data/train-\*
    split: train
  - path: data/val-\*
    split: val
dataset_info:
  dataset_size: 141628052928
  download_size: 75032904415
  features:
  - dtype: string
    name: id
  - dtype: binary
    name: dat
  - dtype: string
    name: hea
  - dtype: image
    name: mask
  - dtype: image
    name: img
  - dtype: binary
    name: dat_T0
  - dtype: string
    name: hea_T0
  - dtype: image
    name: mask_T0
  - dtype: image
    name: img_T0
  splits:
  - name: train
    num_bytes: 134493058267
    num_examples: 4112
  - name: val
    num_bytes: 7134994661
    num_examples: 217
license: apache-2.0
---

# Info

The dataset was generated using this fork of ECG-Image-Kit\
<https://github.com/Ahus-AIM/ecg-image-kit>

and was used to train the segmentation network in\
<https://github.com/Ahus-AIM/Open-ECG-Digitizer>

# Download the dataset

``` python
from datasets import load_dataset

ds = load_dataset("Ahus-AIM/Open-ECG-Digitizer-Development-Dataset")
```

# Mandatory citation

If you use this dataset, please cite

``` bibtex
@article{stenhede_digitizing_2026,
  title        = {Digitizing Paper {ECGs} at Scale: An Open-Source Algorithm for Clinical Research},
  author       = {Stenhede, Elias and Bjørnstad, Agnar Martin and Ranjbar, Arian},
  journal      = {npj Digital Medicine},
  year         = {2026},
  doi          = {10.1038/s41746-025-02327-1},
  url          = {https://doi.org/10.1038/s41746-025-02327-1},
  shorttitle   = {Digitizing Paper {ECGs} at Scale}
}

@article{Shivashankara2024ECGImageKit,
  title   = {ECG-image-kit: A synthetic image generation toolbox to facilitate deep learning-based electrocardiogram digitization},
  author  = {Shivashankara, Kshama Kodthalu and
             Deepanshi and
             Shervedani, Afagh Mehri and
             Reyna, Matthew A. and
             Clifford, Gari D. and
             Sameni, Reza},
  journal = {Physiological Measurement},
  year    = {2024},
  publisher = {IOP Publishing},
  doi     = {10.1088/1361-6579/ad4954}
}
```