TBC-TJU/MetaBCI
MetaBCI: China’s first open-source platform for non-invasive brain computer interface. The project of MetaBCI is led by Prof. Minpeng Xu from Tianjin University, China.
What it solves
MetaBCI is designed to streamline the development of non-invasive brain-computer interfaces (BCIs). It addresses the difficulty of working with fragmented BCI datasets in different formats, the lack of Python implementations for modern decoding algorithms, and the complexity of setting up online BCI experiments.
How it works
The platform is divided into three main modules:
- brainda: Handles dataset importing, EEG data pre-processing, and the implementation of EEG decoding algorithms.
- brainflow: Provides a high-speed framework for online EEG data processing.
- brainstim: A module for designing BCI experiment paradigms (such as MI, P300, and SSVEP).
Who it’s for
Researchers and developers working on brain-computer interfaces who want to focus on algorithm improvement and experimentation rather than data preparation and pipeline setup.
Highlights
- Comprehensive Algorithm Library: Includes decomposition methods (CSP, CCA, TRCA), manifold learning (Riemannian Geometry), and deep learning models (EEGNet, ShallowConvNet).
- Dataset Support: Built-in support for numerous MI and SSVEP datasets.
- Integrated Pipeline: Allows users to bring pre-trained models into an online decoding pipeline.
- ** Lemmas to MOABB**: Enhances MOABB APIs with hook functions for preprocessing and accelerated data loading via joblib.
Related
- Project
- Project
- Project
- Project