Table of contents

• General info
• Setup
• Data preparation
• Analysis guide

General info

Epigame is a project for epileptogenic network definition.

The project is created with:

• Python version: 3.7.1

Setup

Setup is customized for Windows. To run this project, use conda package manager. To install conda, refer to: https://docs.conda.io/en/latest/miniconda.html

Clone the epigame repository to your local machine. Open the Anaconda Prompt. Navigate to the epigame folder and set up the virtual environment:

$ cd epigame
$ conda create --name epigame --file environment.yml

Data preparation

1. Extract the SEEG data of a single seizure as an EDF+ file. The recording should be 10 minutes long, ending at seizure end.
2. Place the EDF+ files in the data folder.

Analysis guide

To analyze the data, run the following command:

$ python main_pipeline.py

You will be asked to input the time window, connectivity method, and the frequency band of interest.

Example: Analysis of transition to seizure by spectral coherence (imaginary part) in low gamma band (30-70 Hz).

Time window:
1. Non-seizure (baseline)
2. Pre-seizure
3. Transition to seizure
4. Seizure
Indicate a number: 3

Connectivity method:
1. PEC
2. Spectral Coherence
3. Phase Lock Value
4. Phase-Lag Index
5. Cross-correlation
6. Phase-amplitude coupling
Indicate a number: 2

Imaginary part (Y/N): Y

Filter the signal (Y/N): Y

Set band range min: 30
Set band range max: 70

The preprocessed epochs are saved in "../results/preprocessed". The identified epileptogenic networks are saved in "../results/EN".