Sustained thought can control robotic arms

Recent scientific breakthrough makes it more viable for sustained use of thoughts to control robotic arms, giving hope to paralysed people

Andy Ho, Straits Times 26 May 2012

NEW study just published in the journal Nature reports on a 58-year-old woman and a 66-year-old man, both with longstanding paralysis, who are now able to move a robotic arm using pure thought.

Their brains have been implanted with hair-thin electrodes comprising a brain-machine interface. This device interfaces the patient’s brain to a computer. The interface can pick up thoughts. That’s not as fanciful as it sounds, since thoughts or brain activity generate electrical signals. These are passed on to a computer which reads and interprets these signals to offer control of an external object or device.

The woman, who was paralysed in all four limbs after a stroke, had such an interface implanted in her brain five years ago. At that time, she had already been paralysed for 10 years.

So the latest success as reported suggests that the brain region called the motor cortex involved can still function after it has not been controlling any limb movement for 15 years.

Note that this is a very crucial point as the relevant parts of the brain must work after years of paralysis of the limbs, say, if the paralysed person is ever to gain thought control of robotic limbs.

In 2002, University of Pittsburgh scientists showed that temporarily paralysed monkeys with an interface could feed themselves using thought- controlled robotic arms.

In 2007, Brown University scientists reported a young man paralysed from the neck down who had a brain-machine interface that enabled him to move a cursor in a video game using his thoughts.

When he thought ‘go left’, the cursor did just that and when he thought ‘go right’, that’s what the cursor did.

The interface was basically a tiny sensor with 100 hair-thin electrodes implanted in his brain that transmitted his thoughts from the motor cortex or brain region that controls limb movements to a computer that decoded his intentions, whether the cursor was to go left or right.

With the latest report, every necessary step for such an interface to give humans thought-control of robotic arms has now been proven to be, in principle, eminently do-able.

What are these steps?

First, it has now been shown that a brain-machine can still function several years after implantation in the brain. Scientists have known for a decade now that implanted electrodes allow for mental control of an external object or device rigged up to the brain.

But if the implanted electrodes stopped functioning after a while, one would have to re-implant new ones periodically. Implantation of anything into the brain, especially if periodically, is no proposition to be taken lightly. That a brain-machine interface actually has longevity is thus a huge breakthrough.

Second, the brain region involved in limb movement in human beings has now been shown to continue to function after years of disuse. This was an unknown until it was demonstrated to be so.

Lastly, the brain has also been shown to be ‘plastic’ enough to learn a new skill to achieve fine control of robotic limb movements for eating and drinking, say. For example, the 58-year old woman in the latest Nature report used a robotic arm to drink coffee from a bottle. Though reaching for and grasping the bottle with the thought-controlled robotic arm was not as fast or as precise as the natural arm, this was proof in principle that it is possible for people with severe paralysis to reacquire three-dimensional manipulative skills. This promises to make life easier for them to perform many daily things we take for granted, such as eating, drinking, brushing teeth and so on.

In fact, in such situations, the external device may even feel like a natural part of the body.

This is akin to the situation where workers handle real things at a remote site by using robotic arms. These include handling radioactive material in a nuclear plant or manipulating objects in a hazardous undersea environment.

When fully immersed in such work, operators report forgetting that they are not actually at the remote site. The technology recedes from the consciousness and they report feeling as though they are right there where the robotic arms are.

Some operators report identifying with the robotic arms so completely that they deal with the remote situation as fluidly and as fluently as if they were working with their own arms.

Some may ask why scientists don’t try to use such interfaces to return to patients actual control of their own disused limbs instead of clunky robotic ones. The answer is that paralysed muscles which have not been getting nerve signals from the brain for a while become dysfunctional as a result.

Depending on the level of the spinal cord at which the injury occurred, the muscles in question will develop either involuntary jerky contractions or they will simply atrophy or waste away. There is currently no way to reverse either set of problems.

So there is little point in trying to re-achieve control of a limb that contracts spasmodically on its own or one that is wasted. For now, trying to get the paralysed to achieve fine control over robotic limbs is more promising.

These limbs could be designed as an exoskeleton to be slipped over the person’s own limbs. An exoskeleton is an external shell of soft fabric equipped with smart motors and intelligent joints.

Thought-controlled exoskeletal upper limbs could enable patients to handle everyday things. Similar lower limbs could enable them to stand up and walk.

But a major hurdle remains: the risk of stroke in some patients if they wore an exoskeleton. This is especially so if the spinal cord injury was at or above the mid-chest level, especially near the neck.

Such an injury cuts off the specialised nervous system that is involved in flight-or-fight. Because that system is cut off, a tight exoskeleton stimulating the muscles and/or skin below that injury level could trigger a huge spike in blood pressure that leads to bleeding inside the brain, which would be a devastating stroke.

In any case, the brain-computer interface has not been perfected yet. One significant problem is that neural tissue form around the implanted electrodes after a decade or so. This causes the electrical signals they pick up from the brain to degrade in quality over time. This means periodic re-implants into the brain, not a prospect one looks forward to.

Much work remains to be done. To date, no one anywhere has a mobile working unit. Reports of successful thought-controlled manoeuvres have taken place only in laboratory experiments. But the rapid advent of such news tells us that Ironman is more likely by the year

andyho@sph.com.sg

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: