Skip to content

Step 3: BIDS-Conversion (from .edf file)

Now the actual BIDS-conversion happens. The ANC implemented a command line interface (CLI) pipeline based on MNE-BIDS that can be run locally and automatically converts and uploads the converted data in a single merge request.

Requirements (One-time setup of the Conversion Pipeline)

Besides requirement to the actual EEG data, some technical requirements need to be fulfilled in order to convert EEG-Data into the BIDS format.

1.) Download and install Python locally. Please use a Python version 3.10 or above.

2.) Create a virtual environment and activate it:

  • Windows: Press Win + R, type cmd and hit Enter
  • Linux: open a Terminal
#Windows
python -m venv name-of-your-environment
name-of-your-environment\Scripts\activate

#Linux
python3 -m venv name-of-your-environment
source name-of-your-environment/bin/activate

Note

if the activation of the virtual environment does not succeed on a Windows OS, try call name-of-your-environment\Scripts\activate.

When the above steps suceed you now see (name-of-your-environment) in the CommandLine. This is a requirement for the next step! Do not proceed in case of a different behavior - feel free to reach out to any ANC Data Steward for help.

3.) Install the EEG Conversion Tool via 

pip install git+https://gitlab.com/ccns/neurocog/neurodataops/anc/pipelines/eeg_conversion.git
You should now see Successfully installed eeg_conversion

4.) Create a Personal Access Token and store it safely

5.) Create a config file

The EEG Conversion pipeline automatically uploads the data to the respective repository. In order to do so, the authentication credentials (i.e. the Personal Access Tokes) has to be available to the conversion pipeline locally.

  • Windows: Create a .python-gitlab.cfg file in C:\Users\<YourUsername>
  • Linux: Create a .python-gitlab.cfg file in your home directory (~/)
  • Open it and add the following content:
[eeg-acquisition]
url = https://data.anc.plus.ac.at/
private_token = <add-your-personal-access-token-here>

Congratulations - the ANC EEG Conversion Pipeline is ready to use 🎉

Run the EEG Conversion

1.) Open a command line:

  • Windows: Press Win + R, type cmd and hit Enter
  • Linux: open a Terminal

2.) Make sure that your virtual environment is activated (you can see it via the (name-of-your-environment) at the beginning of the CommandLine). If it's not activated, run the following command:

#Windows
name-of-your-environment\Scripts\activate

#Linux
source name-of-your-environment/bin/activate

3.) Run the pipeline

The command for running the pipeline:

#Windows
eeg_conversion ^
-f group_project_subject_session.edf ^
-o path\to\output_dir ^
-m path\to\eeg_metadata.csv ^
-t rest ^
-hc 58.5

#Linux
eeg_conversion \
-f group_project_subject_session.edf \
-o path/to/output_dir \
-m path/to/eeg_metadata.csv \
-t rest \
-hc 58.5 \
  • Example usage:

#Windows
eeg_conversion ^
-f C:\Users\YourUsername\Documents\EEG\anc-testing_eeg-demo_S301_2.edf ^
-o C:\Users\YourUsername\Documents\EEG\workdir ^
-m C:\Users\YourUsername\Documents\EEG\metadata.csv ^
-t rest ^
-hc 50

#Linux
eeg_conversion \
-f /home/barbara/Documents/EEG/anc-testing_eeg-demo_S301_2.edf \
-o /home/barbara/Documents/EEG/workdir \
-m /home/barbara/Documents/EEG/metadata.csv
-t rest \
-hc 50

  • Explanation of the Arguments
Argument Short Type Description
--file_path -f str Path to the EEG data file. (Required) \n Please stick to the file naming conventions outlined in Step 1
--output_dir -o str Output directory for the converted BIDS data. Please indicate a separate directory and do not rely on the default. (Default: current directory) This directory serves as a temporary storage for the converted data - it is removed after the data prepconvload succeeded.
--metadata -m str Path to a CSV file containing additional EEG metadata. (Required) Please refer to the instructions for the metadata.csv in Step 2
--task -t str Task name associated with the EEG data. (Required)
--headcircumference -hc float Head circumference of the subject in cm. (Required)

Congratulations 🎉🎉🎉🎉 Your EEG data should have arrived in your ANC Dataset project - please continue with Step 4!