Data and Codes


MF BCI-HMD Data (zip file, 1.09 GB)

  • Scalp‑recorded EEG data were collected during a head‑mounted display (HMD) virtual‑reality brain–computer interface (BCI) experiment designed to examine mental fatigue (MF) induced by prolonged motor‑imagery tasks.
  • Resting state EEG with the eyes-closed and eyes-open conditions was recorded before and after each training session.
  • Each BCI-VR training session consists of three approximtely 20 min long blocks of motor imagery trials.
  • Preliminary analysis was presented the IEEE International Conference on Systems, Man, and Cybernetics (SMC), see Evetovic et al, 2025.
  • Final analysis and the derived mental fartigue model were published in Evetovic et al, 2026.
  • The description of the experiment and the EEG data technical details can be downloaded here as a PDF file. Please contact us if you need any additional information.
    • This data and analysis were carried out with the support of the EU NextGenerationEU through the Recovery and Resilience Plan for Slovakia under project No. 09I03‑03‑V04‑00443.

BCI-VR Data : scalp-recorded EEG data of the brain-computer interface (BCI) in a virtual reality (VR) experiment


  • BCI-VR data were recorded during neurorehabilitation training of post-stroke subjects in VR.
  • Resting state EEG with the eyes-closed and eyes-open conditions was recorded before and after each training session.
  • Each BCI-VR training session consists of three blocks of training, each with ten motor imagery (MI) trials.
  • To discriminate between resting state and MI periods, we used time-scores of a PARAFAC atom, see our paper.
  • Preliminary analysis was presented the the 17th International Work-Conference on Artificial Neural Networks, (IWANN2023), see our talk.
  • Description of the experiment and EEG data technicalities can be downloaded here as a pdf file. Please contact us if further information is needed.
Subject 201
  • A 66-year-old man with residual right-sided hemiparesis after a stroke.
  • Data were recorded from October 13, 2022, to December 16, 2022, representing 16 training days.
  • BCI-VR Data
  • Resting state EC and EO Data
Subject 207
  • An 84-year-old man with residual left-sided hemiparesis after a stroke.
  • Data were recorded from May 11, 2023, to June 22, 2023, representing 13 training days.
  • Before the training, after each training block, and after the training, a collection of questions focused on mental state and cybersickness was collected. The Excel document with the answers is linked here.
  • BCI-VR Data
  • Resting state EC and EO Data



NBO Data (zip file, 46 MB): simulated narrow band scalp-recorded EEG oscillations

  • We applied an anatomical forward model consisting of 2,004 dipoles placed in gray matter. This model was obtained from the on-line repository http://mikexcohen.com/data/ (downloaded, June 20, 2020).

    We modeled narrowband oscillatory sources with seven dipoles at four different frequencies
    • θ rhythm at 5 Hz, a dipole located at the frontal midline region
    • μ rhythm at 8 Hz, left and right hemisphere dipoles located within the somatosensory cortex
    • β rhythm at 14 Hz, left and right hemisphere dipoles located within the somatomotor cortex
    • α rhythm at 11 Hz, left and right hemisphere dipoles located within the posterior cortex
  • More details:

    Rosipal R., Rostakova Z., Trejo L.J.
    Tensor Decomposition of Human Narrowband Oscillatory Brain Activity in Frequency, Space and Time. pdf document
    Biological Psychology, 169, 108287, 2022.
    abstract

Other Data

  • If you are interested in any dataset published in our studies, please contact us.
Kernel PLS MATLAB code

  • Kernel-based methods - kernel PLS, kernel SIMPLS, kernel PCR, kernel RR, EM kernel PCA
    please send an e-mail to roman.rosipal@savba.sk