Brain Pipe

Preprocess brain imaging datasets in a fast and re-usable way.

Motivation

This repository contains code to efficiently preprocess brain imaging datasets in python3, predominantly for machine learning downstream tasks.

The initial goal of this code was to preprocess the public EEG dataset of the ICASSP 2023 Auditory EEG challenge called SparrKULee.

Parts of this code were translated from our internal matlab codebase at ExpORL, maintained primarily by Jonas Vanthornhout and Marlies Gillis

Installation

You can install this repository as a pip package.

pip install brain_pipe

How to use

You can write your own preprocessing script, see the docs, or you can use the cli with configuration files to preprocess your data.

brain_pipe config_file.extension
# e.g. brain_pipe sparrKULee.yaml

Requirements

Python >= 3.8

Contributing

Contributions are welcome. Please open an issue or a pull request. For more information, see CONTRIBUTING.md. This package is created and maintained by ExpORL, KU Leuven, Belgium.

Example usage and starting guide

Read the docs for a more detailed explanation of the pipeline.

For a simple example, see the docs. For a more elaborate example, see examples/exporl/sparrKULee.py