A mobile brain-body imaging dataset recorded during treadmill walking with a brain-computer interface.

Area of Science:

• Neuroscience
• Biomedical Engineering
• Human-Computer Interaction

Background:

• Mobile brain-body imaging (MoBI) integrates brain activity recording with full-body motion capture.
• Brain-computer interfaces (BCI) offer potential for controlling external devices using neural signals.
• Understanding brain-body dynamics during locomotion is crucial for developing advanced assistive technologies.

Purpose of the Study:

• To present the first published MoBI dataset collected during treadmill walking.
• To investigate neural correlates of gait control within a closed-loop BCI paradigm.
• To provide a resource for optimizing BCI decoders for real-time locomotion tasks.

Main Methods:

• Acquisition of synchronized 60-channel scalp electroencephalography (EEG) and lower limb joint angles (goniometers) during treadmill walking.
• Inclusion of electrooculogram (EOG) and EEG impedance for artifact correction and source localization.
• Experimental design with three conditions: standing, treadmill walking, and BCI-controlled treadmill walking.

Main Results:

• The dataset contains synchronized brain and lower limb kinematic data from eight healthy subjects across multiple trials and conditions.
• The closed-loop BCI condition allowed subjects to control a virtual avatar's gait in real-time using their EEG signals.
• Data preprocessing steps included artifact removal and channel localization for robust analysis.

Conclusions:

• This unique MoBI dataset during walking provides a valuable resource for studying brain-gait interactions.
• It enables research into how BCIs influence neural activity during locomotion.
• The dataset can facilitate the development and optimization of BCI systems for gait rehabilitation and control.