Skip to content

The repository of the source code of "Correlating Electrocardiograms with Echocardiographic Parameters in Hemodynamically-Significant Aortic Regurgitation Using Deep Learning"

License

Notifications You must be signed in to change notification settings

Urania880519/AR-detection-with-ECG

Repository files navigation

About

The source code of "Correlating Electrocardiograms with Echocardiographic Parameters in Hemodynamically-Significant Aortic Regurgitation Using Deep Learning"

Summary_r1

Files

The dataset is composed of a mixtutre of Japanese and Taiwanese data

  • Model training and validation: main.py
  • Dataset preperation: dataset.py
  • Testing code: test.py
  • Shapley value calculation and visualization: SHAP.py

We also provide the version of code to train the model only on Taiwanese dataset and finetune on Japanese dataset

  • Please refer to the transfer_learning.sh

Models

  • Base: models weights trained solely on Taiwanese dataset
  • MixedData: models trained on Taiwanese+Japanese dataset
  • Finetuned: models finetuned on the Japanese dataset

Citation

Li YT, Chiang KC, Shieh AT, et al. Correlating Electrocardiograms with Echocardiographic Parameters in Hemodynamically-Significant Aortic Regurgitation Using Deep Learning. Acta Cardiol Sin. 2024;40(6):762-780. doi:10.6515/ACS.202411_40(6).20240918B

About

The repository of the source code of "Correlating Electrocardiograms with Echocardiographic Parameters in Hemodynamically-Significant Aortic Regurgitation Using Deep Learning"

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published