See our Pathology Detector space at:

Or run the code in Binder:

APD EEG

Code for "Amplifying pathological detection in EEG signaling pathways through cross-dataset transfer learning"

Run the Jupyter notebooks

You only need to open and run the following Jupyter notebooks in order. No need to set up any enviroment or download any datasets. We use a mock version of TUAB and NMT for these notebooks, however, the code to download these datasets is also available in the first notebook.

1. Run "s1_data_preparation.ipynb" to prepare the data.
2. Run "s2_supervised_learning.ipynb" to train the model.
3. Run "s3_transfer_learning.ipynb" to fine-tune the model.

If you want to create an environment to run the code, you can use the following command:

Setup the environment

You can create a new environment using the provided environment.yaml file. .. code-block:: bash

conda env create -f environment.yaml

# Activate the environment conda activate APD_EEG

Citation

If you find this code useful, please cite our paper:

@article{darvishi2024amplifying,
  title={Amplifying pathological detection in EEG signaling pathways through cross-dataset transfer learning},
  author={Darvishi-Bayazi, Mohammad-Javad and Ghaemi, Mohammad Sajjad and Lesort, Timothee and Arefin, Md Rifat and Faubert, Jocelyn and Rish, Irina},
  journal={Computers in Biology and Medicine},
  volume={169},
  pages={107893},
  year={2024},
  publisher={Elsevier}
}