/DL-EEG

Primary LanguageJupyter Notebook

Github repo supporting paper

Deep Convolutional Neural Network Applied to Electroencephalography: Raw Data vs Spectral Features

(Published paper: https://ieeexplore.ieee.org/document/9630708)

Code layout

  • environment.yml: conda environment set up
  • HBN_NEMAR_Pheno.csv: Subjects information
  • GSN_HydroCel_129.sfp: Channels standard position file provided by HBN
  • restingstate_prepare_clean_master.m: Preprocess data
  • load_data_master.m: Take result from restingstate_prepare_clean_master.m and sub-select channel to prepare final raw and spectral data.
  • SexPrediction-Original-Raw.ipynb: Notebook to train referenced work model on raw data (R-SCNN)
  • SexPrediction-Original-Topo.ipynb: Notebook to train referenced work model on spectral data (S-SCNN)
  • SexPrediction-VGG-Raw.ipynb: Notebook for training repurposed VGG-16 model on raw data (R-VGG)
  • SexPrediction-VGG-Topo.ipynb: Notebook for training repurposed VGG-16 model on spectral data (S-VGG)
  • utils.py: supporting classes and functions for the Jupyter notebooks