Scientific Progress in recent years has successfully shown that, in principle, it is feasible to drive prostheses or computers using brain activity: Monkeys have learned to move a computer cursor or a robotic arm [1, 2]. People have learned to use simple prototypes to write short texts or to control technical devices [3, 4].

The focus of worldwide research in this new technology, known as brain-machine-, or brain-computer interface, has been based on two different prototypes: Non-invasive BMIs, which measure activity from large groups of neurons with electrodes placed on the surface of the scalp (EEG), and invasive BMIs, which measure activity from single neurons with miniature wires placed inside the brain.
We are investigating a new minimally invasive approach based on electrodes placed on the surface of the brain. This technology minimizes medical risks while still providing high signal quality.

The General principle underlying Brain-Machine Interfaces

In healthy subjects the primary motor area of the brain sends movement commands to the muscles via the spinal cord.	In many paralyzed people this pathway is interrupted, i.e. due to a spinal cord injury.	A new treatment being researched: Electrodes measure activity from the brain. A computer based decoder translates this activity into commands for the control of muscles, a prosthesis or a computer.

Sample experiment performed in Freiburg: Recording of human brain activity during arm movement

Here you can see a patient in the university clinic who is performing hand movements while we are recording signals from the surface of his cortex (electrocorticogram, ECoG). The patient is actually undergoing a pre-surgical epilepsy diagnosis and has voluntarily agreed to take part in this study during his diagnosis. The study has been approved by the universities ethics committee. (click on picture to start movie)

Signals recorded from the surface of the brain can be used to decode the movement intentions of a human subject:
Replayed on the left side are the subjects arm movements (a green cursor that had to be moved to targets which appeared randomly). The subjects brain activity during these movements is shown on the right side (voltages recorded at the electrode grid are colour-coded).
The brain signals can be used for a reconstruction of the subjects arm movements, represented by the red dot (seen in the left side, click on picture to start movie).

Weblinks

Other BMI-Labs in Germany

• Berlin Brain-Computer Interface
• Institute for Medical Psychology and Behavioral Neurobiology (University of Tübingen)

Selected BMI-Labs in other countries

• Laboratory of Miguel A.L. Nicolelis (Duke University, USA)
• Motor cortex research (Hebrew University of Jerusalem, Israel)
• Laboratory of Brain-Computer Interfaces (TU Graz, Austria)

Pioneering companies in the USA

• Cyberkinetics
• Neural Signals