Skip to content

okbalefthanded/bci_toolbox

Repository files navigation

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