moabb.datasets.BNCI2014_004#

class moabb.datasets.BNCI2014_004[source]#

BNCI 2014-004 Motor Imagery dataset.

PapersWithCode leaderboard: https://paperswithcode.com/dataset/bnci2014-004-moabb-1

Dataset summary

#Subj

9

#Chan

3

#Classes

2

#Trials / class

360

Trials length

4.5 s

Freq

250 Hz

#Sessions

5

#Runs

1

Total_trials

32400

Participants

  • Population: paralysis

  • Age: 24.6 years

  • Handedness: right

  • BCI experience: naive

Equipment

  • Amplifier: g.tec Guger Technologies OEG (two 16-channel biosignal amplifiers)

  • Electrodes: Ag/AgCl

  • Montage: 10-20

  • Reference: Car

Preprocessing

  • Data state: preprocessed

  • Bandpass filter: 0.5-100 Hz

  • Steps: EOG regression artifact reduction, bandpass filtering, notch filtering

  • Re-reference: car

Experimental Protocol

  • Paradigm: imagery

  • Feedback: none

  • Stimulus: avatar

BCI Competition IV Dataset 2b [1].

Dataset Description

This dataset consists of EEG data from 9 subjects. The cue-based BCI paradigm consisted of two different motor imagery tasks, namely the imagination of movement of the left hand (class 1) and the right hand (class 2). Two sessions on different days were recorded for each subject. Each session is comprised of 6 runs separated by short breaks. One run consists of 20 trials (10 for each of the two possible classes), yielding a total of 120 trials per session.

The subjects were sitting in a comfortable chair in front of a computer screen. At the beginning of a trial (t = 0 s), a fixation cross appeared on the black screen. In addition, a short acoustic warning tone was presented. After two seconds (t = 2 s), a cue in the form of an arrow pointing either to the left or to the right appeared and stayed on the screen for 1.25 s. This prompted the subjects to perform the desired motor imagery task. No feedback was provided. The subjects were ask to carry out the motor imagery task until the fixation cross disappeared from the screen at t = 6 s.

Three bipolar channels (C3, Cz, C4) and three EOG channels were recorded. The signals were sampled at 250 Hz and bandpass-filtered between 0.5 Hz and 100 Hz. The reference was the left mastoid and the ground was the right mastoid. The electrode montage is a reduced version of the 10-20 system.

References

[1]

Tangermann, M., Muller, K.R., Aertsen, A., Birbaumer, N., Braun, C., Brunner, C., Leeb, R., Mehring, C., Miller, K.J., Mueller-Putz, G. and Nolte, G., 2012. Review of the BCI competition IV. Frontiers in neuroscience, 6, p.55.

Notes

Note

BNCI2014_004 was previously named BNCI2014004. BNCI2014004 will be removed in version 1.1.

Added in version 0.4.0.

This dataset is commonly referred to as “BCI Competition IV Dataset 2b”. It is widely used for binary motor imagery classification tasks.

See also

BNCI2014_001

4-class motor imagery (Dataset 2a)

BNCI2014_002

2-class motor imagery with Laplacian derivations