){kind=icon}
This code repository collects the available source code of the representative DL-based models for classifying MI-EEG signals and runs a leaderboard table to fairly compare these models. The provided shell scripts can conveniently evaluate representative models on public MI-EEG datasets. This repository aims to help researchers learn about current state-of-the-art models and evaluate their proposed models quickly. More summarization and discussions about the representative models can be found in our survey paper (Link). This repository is updated regularly to contain the latest available MI-EEG decoding models.
git clone https://github.com/Henrywang621/DL-based-MI-EEG-models.git
cd DL-based-MI-EEG-models
conda env create -f tf-gpu.yml
conda env create -f torch37b6.yml
conda env create -f torch37c.yml
wget https://repo.anaconda.com/archive/Anaconda3-2021.05-Linux-x86_64.sh
bash Anaconda3-2021.05-Linux-x86_64.sh
conda init
source ~/.bashrc
If you want to evaluate the collected models on the BCI IV 2a dataset, please use the commands below.
cd BCIIV2a_CrossSubjs
chmod +x ./train.sh
sh train.sh
| methods | title | author | year | source code |
|---|---|---|---|---|
| Mixed LSTM/1DConv | Learning Representations from EEG with Deep Recurrent-Convolutional Neural Networks. [Paper] | Bashivan et al. | 2016 | Code |
| Shallow ConvNet | Deep learning with convolutional neural networks for EEG decoding and visualization. [Paper] | Tibor et al. | 2017 | Code |
| Deep ConvNet | Deep learning with convolutional neural networks for EEG decoding and visualization. [Paper] | Tibor et al. | 2017 | Code |
| EEGNet | EEGNet: a compact convolutional neural network for EEG-based brain–computer interfaces. [Paper] | Lawhern et al. | 2018 | Code |
| An end-to-end model | An end-to-end deep learning approach to MI-EEG signal classification for BCIs. [Paper] | Dose et al. | 2018 | Code |
| Cascade model | Cascade and Parallel Convolutional Recurrent Neural Networks on EEG-based Intention Recognition for Brain Computer Interface. [Paper] | Zhang et al. | 2018 | Code |
| Parallel model | Cascade and Parallel Convolutional Recurrent Neural Networks on EEG-based Intention Recognition for Brain Computer Interface. [Paper] | Zhang et al. | 2018 | Code |
| A LSTM model | Validating Deep Neural Networks for Online Decoding of Motor Imagery Movements from EEG Signals. [Paper] | Tayeb et al. | 2018 | Code |
| pCNN | Validating Deep Neural Networks for Online Decoding of Motor Imagery Movements from EEG Signals. [Paper] | Tayeb et al. | 2019 | Code |
| EEGNet fusion | Fusion Convolutional Neural Network for Cross-Subject EEG Motor Imagery Classification. [Paper] | Roots et al. | 2020 | Code |
| C-LSTM | Data augmentation for self-paced motor imagery classification with C-LSTM. [Paper]] | Freer et al. | 2020 | Code |
| GCRAM | Motor Imagery Classification via Temporal Attention Cues of Graph Embedded EEG Signals. [Paper] | Zhang et al. | 2020 | Code |
| TS-SEFFNet | A Temporal-Spectral-Based Squeeze-and-Excitation Feature Fusion Network for Motor Imagery EEG Decoding. [Paper] | Li et al. | 2021 | Code |
| MIN2Net | MIN2Net: End-to-End Multi-Task Learning for Subject-Independent Motor Imagery EEG Classification. [Paper] | Phairot et al. | 2022 | Code |
| EEG-Transformer | Transformer based Spatial-Temporal Feature Learning for EEG Decoding. [Paper] | Song et al. | 2022 | Code |
@article{WANG2024102738,
title = {An in-depth survey on Deep Learning-based Motor Imagery Electroencephalogram (EEG) classification},
journal = {Artificial Intelligence in Medicine},
volume = {147},
pages = {102738},
year = {2024},
issn = {0933-3657},
doi = {https://doi.org/10.1016/j.artmed.2023.102738},
url = {https://www.sciencedirect.com/science/article/pii/S093336572300252X},
author = {Xianheng Wang and Veronica Liesaputra and Zhaobin Liu and Yi Wang and Zhiyi Huang},