The Future of Prosthetics: Mind-Bending Robotic Arms

Researchers have shown that patients paralyzed from the neck down can move robotic arms with their minds, according to a new report in the journal Nature that documents two cases involving brain-stem stroke victims. Margaret Warner discusses the hopes for the technology with Dr. Leigh Hochberg of Massachusetts General Hospital.

Read the Full Transcript


    Next, how paralyzed patients are using their minds to control robotics.

    Margaret Warner looks at a major advance.


    For the first time, researchers have shown that patients paralyzed from the neck down can manipulate robotic arms with their thoughts. A new report in the journal Nature documents two cases involving victims of brain stem strokes.

    One, 58-year-old quadriplegic Cathy Hutchinson was able to direct a mechanical arm to pick up a bottle of coffee and bring it to her lips. It was the first time she'd been able to drink without assistance in nearly 15 years.

    The so-called BrainGate system relies on a sensor of electrodes implanted in the motor cortex of the brain, which controls movement. Simply put, the patient's thoughts are relayed from the sensor to a computer, which sends instructions to the robotic arm.

    In BrainGate's previous breakthrough six years ago, the patient was able to move a computer cursor with his thoughts.

    And for more on all this, we turn to Dr. Leigh Hochberg, a critical care neurologist at Massachusetts General Hospital. He's co-director of the BrainGate research team, which includes Brown University and the Department of Veterans Affairs, among others.

    And, Dr. Hochberg, thanks for being with us.

    How big an advance. . .

    DR. LEIGH HOCHBERG, Massachusetts General Hospital: Good evening, Margaret.


    How big an advance is this in the sort-long-term goal of trying to restore movement to patients who are paralyzed?


    I think that this is a nice and important step in the field that's evolving as we try for people who are unable to move their arms or their legs to develop a technology that would restore either communication or mobility.


    And how does this BrainGate system actually work — if you can, try to explain in layman's terms — in other words, to transmit the intent of the patient to this artificial limb?


    That's right.

    As you described, the motor cortex, which sits right on the top of the brain, is very important for the control of voluntary movement, for example, the movement to reach out and pick up a coffee cup. There's a small array of microelectrodes that's tapped into that top of the motor cortex. And the brain signals, the neurons, we capture that electrical activity, which is really the language of the nervous system.

    That's passed by some fine wires to a pedestal or a plug that protrudes just above the head, and then during our research sessions, those — that electrical activity is brought down to some computers, and the job of those computers is to decode that neural activity, that is, to decode the intended movement of someone who, for example, may be trying to reach out with their arm and hand.


    Now, the woman in the study, Cathy Hutchinson, hadn't moved any limb in nearly 15 years. Yet, her brain — you captured this — was still sending signals — able to send signals to do so. Was that surprising?


    Yeah, she had had no functional use of her limbs for over 15 years. She had had a little bit of residual movement of her arms, but it was of no function to her.

    It is very encouraging, I think, for neurorehabilitation and for the potential to harness these brain signals going forward that even 15 years after a truly devastating stroke that essentially disconnected a perfectly working brain from a perfectly working body that she was able to think about the movement of her own hand, we were able to record those signals, and she was able to reach out and pick up that thermos of coffee.


    Now, what sorts of patients would be — could be helped by this? In other words, what sorts of injury or damage would this basically transcend or bypass?


    In our ongoing pilot clinical trial of the BrainGate neural interface system, which an investigational medical device, we're first looking at how the system might work in people who are tetraplegic or quadriplegic, that is, unable to move their arms or their legs.

    This technology, though, and the ability to record from ensembles of single brain cells, that is, many different individual brain cells, simultaneously to me suggests some great potential, that we would be able to use those signals for people who may perhaps be not so physically impaired as the people that I was describing and the folks who so kindly helped us in this manuscript.


    But, I mean, the sorts of damage, that's spinal cord injuries. These two had brain stem strokes. What sorts of injury could it help and what sorts couldn't it?


    Exactly right.

    So, for people with spinal cord injury, with brain stem stroke, with ALS, known as amyotrophic lateral sclerosis or more commonly as Lou Gehrig's disease, these are all either injuries or diseases from — the person who has this injury or disease is perfectly awake and alert, able to appreciate everything in their environment, able to think and has every desire to move and to communicate.

    And it's for those types of injuries or diseases that this early initial research is being focused.


    So, in other words, the brain's motor cortex, as it's known, that still has to be intact and functioning?


    Well, we're recording right now from the motor cortex, which is one of many parts of the brain that are involved in the control of movement.

    So, while we're learning a lot — and really we have gotten to this point the result of more than 40 years of publicly funded science and clinical research to understand how this part and other parts of the brain work. But it's quite possible that this same array or other technologies would be able to record from other parts of the brain that are similarly involved in the control of movement.


    Now, this — could this also work for amputees?


    As sponsored by the Department of Veterans Affairs, part of this paper that was published today in Nature, we're testing the feasibility for whether this type of BrainGate recorded signal, that is, the signals that can be harnessed from inside the brain, might be of use for the control of a new and advanced prosthetic limb, such as the DEKA prosthetic limb that was discussed in the paper.


    So how far away are we? How far away are families from this having any practical use?


    That's the critical question, of course.

    And I'm encouraged by the stage of the research that we're at and the progress that we're making. There's still a lot of research left to do. And if I look to the future and I think about somebody like either of the two participants in this paper, both of whom not only can't move their arms and their limbs or their arms and their legs in any functional way, but also can't speak, I'm hoping that this type of technology would first be able to restore communication, for example, to be able to type on a computer screen, later on, perhaps, to be able to control a robotic assistive device.

    But the real dream for the research is to one day reconnect brain to limb and to bring those signals from the brain back down to the arm, to stimulate the peripheral nerves in the arm and to allow that person with paralysis to reach out and pick up that coffee cup again.


    Yeah, using their own muscles.

    Well, Dr. Leigh Hochberg, thank you so much.

    And, online, you can watch more video showing how brains and machines can interact. Find a link on our home page.