moabb.datasets.BNCI2015_004#

class moabb.datasets.BNCI2015_004[source]#

BNCI 2015-004 Mental tasks dataset.

PapersWithCode leaderboard: https://paperswithcode.com/dataset/bnci2015-004-moabb-1

Dataset summary

#Subj

9

#Chan

30

#Classes

5

#Trials / class

80

Trials length

7 s

Freq

256 Hz

#Sessions

2

#Runs

1

Total_trials

7200

Participants

  • Population: SCI and stroke

  • Age: 38.5 (range: 20-57) years

  • BCI experience: naive

Equipment

  • Amplifier: g.tec

  • Electrodes: active electrodes

  • Montage: 10-20

  • Reference: left mastoid

Preprocessing

  • Data state: raw filtered

  • Bandpass filter: 0.5-100 Hz

  • Steps: bandpass filter, notch filter

Data Access

Experimental Protocol

  • Paradigm: imagery

  • Feedback: cue-guided, no online feedback during screening

  • Stimulus: avatar

Dataset from [1].

Dataset Description

This dataset contains EEG data from 9 subjects performing five different mental tasks: mental multiplication, mental letter composing, mental rotation, mental counting, and a baseline task.

References

[1]

Zhang, X., Yao, L., Zhang, Q., Kanhere, S., Sheng, M., & Liu, Y. (2017). A survey on deep learning based brain computer interface: Recent advances and new frontiers. IEEE Transactions on Cognitive and Developmental Systems, 10(2), 145-163.

Notes

Note

BNCI2015_004 was previously named BNCI2015004. BNCI2015004 will be removed in version 1.1.

Added in version 0.4.0.

On this page