Table 1 Comprehensive dataset characteristics and recording parameters.

From: Hierarchical attention enhanced deep learning achieves high precision motor imagery classification in brain computer interfaces

Demographics & configuration

Technical specifications

Parameter

Value

Parameter

Value

Participants (M/F)

15 (8/7)

Sampling frequency

250 Hz

Age range

21–28 years

A/D resolution

24-bit

Mean age ± SD

24.3 ± 3.2 years

Online filtering

1–50 Hz

Handedness (LQ)

Right (> 0.7)

Dynamic range

\(\approx\) 144.5 dB

MMSE score

\(\ge\) 27 points

Electrode impedance

< 5 k\(\Omega\)

BDI score

< 10 points

Reference electrode

Linked mastoids

Total electrodes

64 channels

Motor electrodes

22 channels

Positioning system

Extended 10-20

Ground electrode

AFz

Experimental design

Timing & quality metrics

Parameter

Value

Parameter

Value

MI classes

4 (LH, RH, feet, tongue)

Fixation period

2.0 s

Sessions/participant

6 sessions

Cue presentation

1.0 s

Trials/session

48 trials

MI execution

4.0 s

Trials/class/session

12 trials

Inter-trial interval

2.0 s

Total trials/participant

288 trials

Total trial duration

9.0 s

Total dataset size

4,320 trials

  

Quality criteria met

96.3% (4,160/4,320)

Avg. SNR

18.7 ± 4.2 dB

Class balance

Perfect (25% each)

Artifact rate

3.7%