GeoJAX PhysioNet motor-imagery tutorial data ============================================= File ---- physionet_motor_imagery.npz Upstream dataset ---------------- EEG Motor Movement/Imagery Dataset, version 1.0.0 Gerwin Schalk, PhysioNet, 2009 DOI: https://doi.org/10.13026/C28G6P Dataset page: https://physionet.org/content/eegmmidb/1.0.0/ The upstream recordings were collected with the BCI2000 system described in: Gerwin Schalk, Dennis J. McFarland, Thilo Hinterberger, Niels Birbaumer, and Jonathan R. Wolpaw. "BCI2000: A General-Purpose Brain-Computer Interface (BCI) System." IEEE Transactions on Biomedical Engineering, 51(6):1034-1043, 2004. DOI: https://doi.org/10.1109/TBME.2004.827072 License ------- The upstream data files are licensed under the Open Data Commons Attribution License (ODC-By) v1.0: https://opendatacommons.org/licenses/by/1-0/ This file is a derived subset redistributed under the same attribution terms. GeoJAX's MIT software license does not replace or supersede the dataset license. Derivation ---------- The derivative was prepared from subjects S001 through S005 and unilateral motor-imagery runs R04, R08, and R12. Event T1 denotes imagined left-fist movement and T2 denotes imagined right-fist movement for these runs. For every T1 or T2 annotation, the derivative stores the interval from 0.5 seconds through 3.5 seconds after event onset. The eight retained sensorimotor channels, in file order, are: Fc3., Fc4., C3.., C1.., C2.., C4.., Cp3., Cp4. Signals are stored as float32 microvolts at the original 160 Hz sampling rate. No filtering, covariance estimation, trial rejection, or learned transformation was applied before storage. The executable tutorial performs those steps visibly. The derivative contains 225 trials with shape (225, 8, 480), plus event labels, subject identifiers, run identifiers, channel names, and acquisition metadata. The derivative contains no names or contact information. Subject identifiers are the public pseudonymous identifiers used by the upstream dataset.