Brain-computer interface controlled robotic gait orthosis

Individuals with spinal cord injuries often rely on wheelchairs, which can lead to health problems. This study explores using brain-computer interfaces (BCIs) to control robotic devices that help people walk. The study involved recording brain activity (EEG) from participants as they imagined walking. This data was used to create a model that could predict when the person wanted to walk. The brain-computer interface was connected to a robotic gait orthosis (RoGO), a device that helps support and move the legs. Participants were then able to control the RoGO using their thoughts, allowing them to walk on a treadmill.

• 1
  The EEG prediction model achieved an average offline accuracy of 86.30% across both subjects.
• 2
  The cross-correlation between the computer cues and the BCI-RoGO walking epochs averaged 0.812.
• 3
  There were on average 0.8 false alarms per session and no omissions.