README.md

Instructions

Data processing code

1. ECG_MIT_1D_dataprocessing_125Hz: Process data into 1D 125 Hz train/test (for both inter and intra patient paradigm for Kachuee paper replication
2. ECG_MIT_1D_dataprocessing_360Hz: Process data into 1D 360 Hz train/test (for both inter and intra patient paradigm for Romdhane paper replication
3. ECG_MIT_2D_STFT_dataprocessing_360Hz: Process data into 2D STFT images and label for our proposed method

Model train/test code

1. Kachuee_paper_replication: Train and test Kachuee model for both inter and intra patient paradigm
2. Romdhane_paper_replication: Train and test Romdhane model for both inter and intra patient paradigm
3. STFT_Resnet: Train and test our proposed model for inter patient paradigm

Detail paper of our model

https://arxiv.org/ftp/arxiv/papers/2206/2206.14200.pdf

References

1. Kachuee paper: Kachuee, M., Fazeli, S., & Sarrafzadeh, M. (2018, June). Ecg heartbeat classification: A deep transferable representation. In 2018 IEEE international conference on healthcare informatics (ICHI) (pp. 443-444). IEEE.
2. Romdhane paper: Romdhane, T. F., & Pr, M. A. (2020). Electrocardiogram heartbeat classification based on a deep convolutional neural network and focal loss. Computers in Biology and Medicine, 123, 103866.