BCI based on lower-limb motor imagery and a state machine for walking on a treadmill

Brain-Computer Interfaces (BCI) based on motor imagery can be employed to control devices that assist the movement while promoting neuroplasticity. The present paper shows a lower-limb motor imagery BCI that has been designed to control a treadmill and it also considers the level of attention of the subject for limiting the false activations. This BCI works as a state machine in which different classification models were employed depending on the treadmill status during motion or standing. It was tested with three able-bodied subjects and the average performance in closed-loop was 75.6 ±10.3%.