https://github.com/okbalefthanded/bci_toolbox

Classification toolbox for ERP and SSVEP based BCI data
https://github.com/okbalefthanded/bci_toolbox

benchmarking brain-computer-interface classification eeg erp machine-learning ssvep

Last synced: 5 days ago
JSON representation

Classification toolbox for ERP and SSVEP based BCI data

• Host: GitHub
• URL: https://github.com/okbalefthanded/bci_toolbox
• Owner: okbalefthanded
• License: apache-2.0
• Created: 2018-01-23T18:51:01.000Z (almost 7 years ago)
• Default Branch: master
• Last Pushed: 2021-05-03T23:45:05.000Z (over 3 years ago)
• Last Synced: 2024-05-14T00:18:09.672Z (8 months ago)
• Topics: benchmarking, brain-computer-interface, classification, eeg, erp, machine-learning, ssvep
• Language: MATLAB
• Homepage:
• Size: 6.99 MB
• Stars: 50
• Watchers: 4
• Forks: 27
• Open Issues: 5
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# bci_toolbox (work in Progress)

A benchamark classification toolbox for Brain-Computer Interface EEG data

 Data sets available:

 

 ## ERP Data sets:

  1. BCI Compeition III Challenge 2004 (P300 evoked potentials) (http://www.bbci.de/competition/iii/)

  2. EPFL P300 data set (https://mmspg.epfl.ch/BCI_datasets)

  3. P300 speller with ALS patients (set #8) (http://bnci-horizon-2020.eu/database/data-sets)

  4. LARESI inverted face data set (coming soon)

## SSVEP Data sets:

  1. SSVEP Exoskeleton (https://old.datahub.io/dataset/dataset-ssvep-exoskeleton)

  2. Tsinghua Sampled sinusoidal Joint Frequency-Phase Modulation SSVEP (http://www.thubci.org/en/?a=nr&id=100)

  3. San Diego Square Joint Frequnecy-Phase Modulation SSVEP (ftp://sccn.ucsd.edu/pub/cca_ssvep)

- Processing methods available: 

- - Feature extraction: 

    --- Downsample

    --- Multivariate Linear Regression(MLR)

- - classification : 

  - - - LDA

  - - - Regularized LDA (shrinkage-LDA)

  - - - SWDLA

  - - - SVM (LIBSVM)

  - - - Logistic Regression (LIBLINEAR)

  - - - Random Forest 

  - - - SVM+ 

  - - - Canonical Correlation Analysis based methods : 

               - - - CCA, FilterBank CCA (FBCCA), L1-Multiway CCA, MsetCCA, Individual Template CCA (ITCCA)

  - - - Task-related Component Analysis (TRCA)

  

  

  # Setup

  Run the setup.m script

  

  # Usage

  ## First run

  - Download one of the Datasets (or all) in the list.

  - create a new folder inside the Dataset/epochs.

  - Run one (or all, one by one) scripts related to each dataset in the dataio folder.

  

  ## Regular usage

  - Run the script "define_approach_ERP.m" for ERP data

  - Run the script "define_approach_SSVEP.m" for SSVEP data

  

  # Documentation

  coming soon

  

  # Cite us

  coming soon