Decoding of hand movement velocities in three dimensions from the EEG during continuous movement of the arm

Brain-computer interface (BCI) systems can be used to control limb neuroprostheses in order to restore limb functionality of paralyzed persons. Traditionally, only invasive BMI (brain machine interfaces) are thought to provide an adequate signal-to-noise ratio and bandwidth to control an upper limb neuroprosthesis accurately in a continuous manner with sufficient degrees-of-freedom. This paper supports work which already showed that it is possible to decode the velocity of executed arm/hand movements from electroencephalographic signals but using a different movement paradigm and suppressing eye movements. It was possible to decode movement velocities with a high correlation with actual executed movements. This could pave the way to a new type of prostheses control using a direct link betweeen natural hand movement imagination and prostheses movement.