January 8, 2004 - Using technology, brain power to beat paralysis

---

In the U.S. approximately 11,000 new cases of spinal cord injuries lead to permanent paralysis annually. These expand an existing base of approximately 200,000 partially paralyzed (paraplegic) and almost totally paralyzed (quadriplegic) patients.

Previously, in order to modify this paralytic condition, research was focused on reconstructing motor functions by restoring the connection and function of damaged spinal nerve fibers. This approach yielded some encouraging results, such as limited restoration of limb mobility, in animals. However, restoring complex motor functions such as reaching and grasping remained a major challenge.

Frustrated by this limited success, researcher E. M. Schmidt, in 1980, sought an alternate method for better restoration of motor function. He proposed in an Annals of Biomedical Engineering article a direct interface between healthy cortical or subcortical brain tissue and an artificial activator, bypassing the spinal cord injury. That would allow paralyzed patients to control their limbs or prosthetic devices as they wished.

However, not until between 1999 and 2002 did John K. Chapin and Miguel A. L. Nicolelis at Duke University, along with other researchers, propose the use of brain motor interfaces (BMI), that is, neurons connected to a robotic arm. They began by wiring the brains of macaque monkeys to a computer that tracked the monkey's brain activity as it moved a joystick to a target in order to receive a food reward. At the same time, the computer was also connected to a robotic arm which mimicked the monkey's activity. As the monkeys manipulated the joystick, the wiring from the brain identified the brain patterns the monkeys were using. When researchers disabled the joystick, the monkeys found they could control the robotic arm merely by "thinking" about it. Their brain activity was translated into instructions which directed the robot.

Following this dramatic research, M.A. Sirinivasan, director of the Touch Lab at MIT, proposed that 600 miles away, in Cambridge, Mass., a different computer would produce the same actions simultaneously in a robotic arm by use of signals sent over the Internet. Thus, by converting electrical brain patterns into instructions to direct the robotic arm, two robotic arms could act in concert, at the same time with exactly the same movements, 600 miles apart.

This suggests the possibility of transmitting the instructions without direct wiring between the brain and a robotic or prosthetic device, a truly revolutionary advance for these paralyzed patients.

Nicolelis and Chapin envision the following scenario: A neurochip implanted in the skull transmits the brain activity ("thoughts") by way of wireless radiofrequency signals to a small, battery-powered computer attached to the wheelchair. As the patient thinks about the action desired, such as reaching for and grasping an object, these signals are then sent, again wirelessly, to another microchip imbedded in the patient's arm. The signals "instruct" the muscles in the arm to move so as to complete the activity.

In case of inability to use the patient's actual arm, the signals could be transmitted to a small robotic device which would then perform the activity. If necessary, these signals could also manipulate the motorized wheelchair to position it where the patient desires.

Wolf, also at Duke, has already built a prototype neurochip and backpack as described above. Despite the remaining hurdles to accomplish the operation of the first human neuroprosthesis, there is reason to be optimistic that such an event will be accomplished within this decade.