basil-ssvep-online

The BASIL BCI project: SSVEP on-line classification

Structure

• figures - containes stimuli pictures that are displayed in the desktop application
• src - Python source codes of the main project
• psychopy - contains Psychopy scripts for stimulation (needs to run simultaneously with the classifier)
• requirements.txt - define Python dependencies of the project

Running the classifier

Requires Python 3.7.

To execute:

1. Directly run run_basil_app.bat to install all dependencies and run the main application.
2. If all dependencies including this project are installed, go to ./src and execute python main.py.

Quick start

• Once you run the application, notice two conditions that have to be met before classification can start. Both EEG data and marker-related LSL streams must be visible (i.e. they must turn from OFF to ON in the GUI). To satisfy these conditions:

1) Run Psychopy stimulation (Run ./psychopy/ssvep_basil.py). 2) Ensure active EEG LSL stream (such as BASIL stream) is on, and its name corresponds to eeg_stream_name variable in: src/data/config.py

• Now the Start button should be enabled and clicked on. The participant gets ready for stimulation (in each run, he/she selects one picture to focus on).

• After the stimulation, both a winning picture, classification confidence and a spectral plot are shown. As an optional feedback, the user can click on True or False buttons.

• When the marker stimulation is over or the user request stop using the Stop button, the application execution terminates after LSL timeout (to be set in src/data/config.py ).

• Then, it can be started again once both streams are available.