Modelpublic

fmenegui/pretrained_image_ecg_doc

sourceHugging Faceupdated 1y agoView on Hugging Face

0likes

Model Card

Pretrained Image-Based ECG Model (ConvNeXt-Tiny)

This repository contains a pretrained ConvNeXt-Tiny model for image-based 12-lead ECG classification. The model was trained using a combination of the following datasets:

—CODE15 [1]
—InCor-AMB [2]
—CINC21 [3]

The checkpoint model_fold0.ckpt can be loaded using the example provided in example.py.

References

[1] Ribeiro, A. L. P., et al. Telehealth Platform for Multicenter ECG Analysis in Brazil: CODE Study.

[2] Dias, F. M., Ribeiro, E., Moreno, R. A., Ribeiro, A. H., Samesima, N., Pastore, C. A., Krieger, J. E., Gutierrez, M. A. Artificial intelligence-driven screening system for rapid image-based classification of 12-lead ECG exams: A promising solution for emergency room prioritization. IEEE Access, vol. 11, pp. 121739–121752, 2023. https://doi.org/10.1109/ACCESS.2023.3316861

[3] Reyna, M. A., Sadr, N., Perez Alday, E. A., Gu, A., Shah, A. J., Robichaux, C., et al. Will Two Do? Varying Dimensions in Electrocardiography: The PhysioNet/Computing in Cardiology Challenge 2021. Computing in Cardiology, vol. 48, pp. 1–4, 2021.