The Future of Mind-Controlled Prostheses

The future of high-tech prostheses and the clinical benefits they provide emerges through research.

© 2009

It was less than a decade ago that the story of Jesse Sullivan seemed incomprehensible; out of the realm of imagination. Nearly every newspaper story, and all newspapers seemed to write one on Sullivan, used the words science fiction to describe it.

If they did not call it science fiction, then “cyborg” or “bionic man” were high on the list of descriptive phrases used by the authors. At the time – and in many ways still – the wonderment was perfectly understandable and not a bit of hyperbole.

Brain signals

It was May of 2001, when Sullivan, a high-power lineman, was electrocuted so badly that both of his arms were amputated at the shoulder because of severe electrical burns.

Sullivan was fitted with traditional prostheses, and for a time, they worked well enough. Soon, however, the skin grafts used to close his wounds developed sores that caused him unbearable pain and rendered the prostheses all but useless.

The obstacle stirred some creative thinking amongst the minds at the Rehabilitation Institute of Chicago (RIC), where Sullivan was a patient.

A team took nerve endings from Sullivan’s shoulders, grafted them into his chest and gave them time to grow. Surface electrodes placed on the newly-grown muscles in his chest picked up the brain signals being sent to the arm that no longer existed and rerouted them to the prostheses.

In essence, they created shoulder nerves in his chest, and then taught them how to communicate with the prostheses. Think of it as a high-science jury-rigging of the nervous system to bypass the amputated limbs.

A legend born

After training, Sullivan could think “open hand” and his motorized arm would comply. Soon he was completing basic chores. The bionic man and his legend were born.

So why revisit a well-worn story that many already know well?

Because the latest technology – and the breathtaking speed at which it is advancing – threaten to make the miraculous adjectives used to describe Sullivan’s transformation sound mute in comparison.

While the work pioneered by the RIC remains at the cutting edge with value to the patients lucky enough to undergo the procedures, new research takes the concept of thought-controlled prostheses to thrilling dimensions.

Think of it as a high-science jury-rigging of the nervous system to bypass the amputated limbs.

What if, instead of using surface electrodes, the next generation uses electrodes implanted directly in the muscle, targeting muscles much more effectively and streamlining the prosthetic control? What if researchers are close to getting the “bionic” technology to work for lower extremities?

Those “what ifs” may be closer to reality than you think, according to Hugh Herr, PhD, director of the Biomechatronics Group at the Massachusetts Institute of Technology.

“I would expect within a decade,” to see widespread, practical applications, Herr said.

Dreaming even wilder, what if, instead of training the movement of the prostheses through grafting and muscle reinnervation, the brain could exert direct control over the limb? What if it could all be done with an implant the size of a hair?

The last is the version of the future at which the researchers at the University of Pittsburgh are offering a glimpse. In a project they’ve been building incrementally for years, researchers last summer unveiled a pair of macaque monkeys that used their brains to control a robotic arm. With a little training, they could feed themselves and self-correct their motions.

But of course, the future of brain-machine interfaces is not in monkeys rewarding themselves with a tasty treat. The ever-rapidly approaching horizon is using the technology for humans to control sophisticated prostheses.

The ultimate breakthrough in the field is not around the corner or in tomorrow’s newspaper. But Herr, at least, said it’s not the talk that should be relegated to science fiction. In fact, he says, it’s a future from which many current amputees may live to see the fruits.

Central vs. peripheral

Quick college biology class review: the central nervous system refers primarily to the brain and spinal cord, the conductor that orchestrates all the activity of our bodies. The peripheral nervous system is the highway that connects the central nervous system to our limbs and organs and allows them all to communicate.

The distinction is important because when talking about using implants as the future of mind control-prostheses, they have different applications.

Implanting chips into the brain, as is the case of the Pittsburgh monkeys, taps into the central nervous system.

The electrode based-implantations – burying Sullivan’s surface electrodes in the body and creating a wireless connection – deal with the peripheral nervous system

For reasons that are fairly obvious, in humans, implanting into the peripheral system is a method that is farther advanced and closer to practical application.

Within a decade, as the technology becomes more stable, Herr said he could see things moving to an all-implant system, using a wireless connection to communicate with the external prostheses.

“Where right now there are surface electrodes, instead of those being there it will simply be an implant there,” Herr said.

When implanting into the peripheral nervous system, the procedure does not have to be terribly difficult. It is a fairly straightforward surgical procedure that can be done as an outpatient. The incision would be relatively minor.

“I don’t see any fundamental reason why the cost has to be outlandish,” Herr said.

Once implanted, Herr said, the advantages compared to surface electrodes are in terms of toning and being able to target individual muscles.

The chief, if obvious, downside to implants is their invasiveness. If something is to go haywire, they can not be tinkered with on the outside, as in the surface electrodes employed by Sullivan’s doctors.

And as the technology is relatively new, the way the implants will behave over the long term is unknown.

“I would say a dominant concern is long-term viability,” Herr said.

That, combined with the greater risk, are reasons why central nervous system implantations are farther away from widespread application, and in his opinion may not be as good a choice all but for the most serious cases, such as spinal cord injuries or extreme upper-arm amputations when movement is drastically curtailed.

“The central implants are fairly major and somewhat dangerous,” Herr said.

In other words, the benefits must be considerable enough to outweigh the risk when it comes to toying with the central nervous system.

“[In] most cases it doesn’t make sense given the risk of putting the implant into the body…” Herr said. “That’s just my opinion.”

Additionally, Andrew Schwartz, PhD, a professor of neurobiology at the University of Pittsburgh, reminds practitioners of the complexity involved in the perpheral control of prosthetics.

“The assumption is made that these signals are equivalent to those sent to muscles and that if muscle activity is known, then it is easy to relate this to movement of the arm and hand,” Schwartz said. “This is not the case. The relation between muscle activity and limb displacement is complex and we don’t understand it very well.”

Robot trials

While central nervous system implantation may remain too risky for immediate and widespread trial in humans, it’s going quite swimmingly for monkeys on the campus of the University of Pittsburgh.

fIf this sounds vaguely familiar, it’s because the project and similar ones have been ongoing for several years. The monumental progress comes in increments, the latest dose dished out in a paper published in the journal Nature last summer.

According to the paper, written by Schwartz the monkeys practiced control of the robotic arm with a joystick. The arm looked like a large cylinder with elbow and shoulder joints and two claw-like fingers. Researchers pinned back the monkeys own arms to encourage them to use the robotic device.

Sometime after the familiarization period, the researchers implanted a grid of electrodes on the portion of the monkeys’ brain that controlled upper limb movements. After several days of training by the team, the monkeys were able to reach out and grab grapes and other snacks dangling in front of them.

The study called the movement “one natural motion.”

The difference between this and earlier experiments, Schwartz wrote, was the real-world functionality. Prior experiments at his lab involved monkeys using their minds to control an arm moving a cursor on a computer screen.

In this go-around, the monkeys were able to transfer the food to their mouths greater than half the time and closer to two-thirds. In prior experiments where feeding was involved, the success ratio was much less.

Perhaps even more significant was the refinement of the motion. The monkeys showed the ability to adapt their movements to unforeseen outside circumstances, such as using the finger to push the food further in the mouth or licking the remnants of food from the fingers.

The implication being that the technology may someday be headed toward more seamless, life-like control. As Herr cautioned, however, this type of implementation is not yet practical in human practice. This won’t be in the local clinic anytime soon.

Not only would the implantation be risky, a human could hardly carry around all the trappings of Schwartz’s experiment. The arm is bulky, requires a lab tech, and was connected to a large computer.

As far as it’s come in such little time, the burgeoning field of neuroprosthetics has a way to go. That’s not the point, however.

Peeling back the future far enough, the potential ramifications for paralyzed individuals – the researchers primary target – and amputees emerge over the horizon. If not quite tangible, they certainly make it easy to imagine a far off future where the technology has clinical benefits.

Tim McManus is a correspondent for O&P Business News.

Leave a Reply

Your email address will not be published.