Hollywood's "Bionic Man" may still be more fantasy than reality, but advances in medical technology are beginning to meld man and machine, according to one biomedical engineer.

Researchers surveyed the current and future capacity of machines that can interact with the brain and nerve cells to help restore some function to people with lost or paralyzed limbs, including victims of spinal cord injury, amputation and stroke.

One current area of research uses brain-machine interfaces, which attempt to interpret commands from the brain and initiate a response by a computer-controlled prosthesis.

In one experiment, researchers have fitted severely paralyzed patients -- who can't even blink their eyes to communicate -- with a brain-implanted chip that they can direct to select letters on a computer screen and thus restore interaction with the outside world.

Other devices circumvent the brain and use functional electrical stimulation to communicate directly with muscles and nerves. Some are commercially available, including the FDA-approved FreeHand, which restores grasping ability in patients with weakened arms who use shoulder movements to trigger radio waves that in turn activate electrodes in their forearms to move.

Another device currently under development is the RoboWalker, which allow patients who have lost motor skills in their legs to walk. The patients use their hands to send messages to their leg muscles, which are stimulated by implanted electrodes.

One of the many obstacles in bionics is that the electronic devices needed to receive the message are often too large to fit inside the brain. "It's a problem of the size of the hardware," Craelius said. "Presently, the brain is not big enough to contain all the power you would need to do things."

However, as long as Moore's law continues to hold true, which states that the number of transistors on a computer chip will double every 18 months, sufficient miniaturization could be reached in a decade.

Another barrier to bionics is the difficulty of integrating the prosthesis with the body's natural tissue. Trying to make a decent place where skin can merge with an artificial limb are decades away from being solved. When you're talking about living cells that have to grow over it, that's a totally difficult magnitude of problem.

Science February 8, 2002;295:1018-1021