Paralyzed Woman Uses Mind-Controlled Robot Arm

MrSeb writes "Using BrainGate, the world's most advanced brain-computer interface, a woman with quadriplegia has used a mind-controlled robot arm to serve herself coffee — an act she hasn't been able to perform for 15 years. BrainGate, which is being developed by a team of American neuroscientists from Brown and Stanford universities, and is currently undergoing clinical trial, requires a computer chip to be implanted in the motor cortex of the patient, which it then transmits to a computer for processing. Like all brain-computer interfaces, the user must train the software — but once this is done, you simply think of a movement, and the software moves the robot accordingly. Moving forward, the researchers would like to miniaturize the system and make it wireless — at the moment, BrainGate users have a box attached to their head, and they're tethered to a computer — which is OK for robot arm use at home, but obviously doesn't grant much mobility. The work was partly funded by DARPA, with the hope of creating more advanced prosthetics for wounded war veterans." This comes on the heels of a 71-year-old man regaining motor function in his fingers after doctors rewired his nerves to bypass the damaged ones.

Comments

Hollywood taught me this!

[durrr]

This will end with a mind controlling robot army.

Re:Hollywood taught me this!

[geekoid]

Why don't you do that?

She Never Asked For This

[IndigoParadox]

"The work was partly funded by DARPA, with the hope of creating more advanced prosthetics for wounded war veterans."

Yeah, sure, that's what it's for, Mr. Sarin.

braingate2

[itamihn]

"Moving forward, the researchers would like to miniaturize the system and make it wireless"

And that is the purpose of Braingate2: http://www.braingate2.org/

20+ years ago...

[Brad1138]

20+ years ago, I saw an episode of a show "Beyond 2000". They had someone hooked up to electrodes of some sort, there was nothing implanted. She was controlling an avatar, in first person, in a 3D environment with her thoughts. I always thought if we could do that 20 years ago why, has it taken so long to see anything more. Even now, this really isn't any more impressive. If we could even control a mouse pointer and "click" with our mind it could be very useful. It wouldn't be much more to then be able to text with our minds, which then opens up a form of telepathy.

I just wonder why this tech seems to have been stuck in its tracks for so long.

Re:20+ years ago...

[virgnarus]

There's a lot of obstacles that need to be overcome that would ultimately explain the slow progression. First of all, the brain is still being understood, and continues to surprise scientists with how it operates. In addition to this, you also have the deal with the body's reaction to invasive brain implants like this one, which they are attempting to avoid through use of more bio-friendly materials to substitute the gold and silicon. There's a lot of studies being conducted on attempting to workaround this through less invasive or noninvasive interfaces (like the consumer brand Emotiv product), but signal output at these levels is very noisy and vague. Don't forget that research is fragmented on this as there's several ways to achieve the same goal, such as either a replacement limbs that moves on neural stimuli, or the use of an external robotic arm as seen here.

Intel is now just getting serious about putting heavy money towards brain interface research, so it's good a big microchip computer like them are putting their vested interest in this, and that will indeed expedite the process, but there's just so much that's still left to be learned and all of it must come together in an affordable package. It's no different than quantum computers - researchers are learning more on the mechanics behind it while at the same time creating working products on what they've learned so far. Things are going to start rudimentary, yes, but one needs to be positive on this that there's actually progress being done.

Money wasted on decades old approach

[Anonymous Coward]

35 year old work, rehashed because "it's got a computer!!!" The technology has never worked well because electrodes, whether the myocardial skin electrodes or the implanted ones of this new technology, wind up getting a tremendous amount of electrical noise from the area of the elctrode, and the noise is from similar signal electrode signals. To sort out the valid signals from all the noise takes about half a second: digital sampling does not help this, and is in fact counterproductive because it adds its own noise and makes the system take much more power.

You can write a lot more papers and get more people excited by your "computer control", a low pass filter with an adjustable gain does a much better job with much less power consumption and much less error, and this has been true since the invention of the Boston Arm. Now, if they'd invented a new electrode technology, I'd be much more excited. But this is a rehash of decades, even centuries old work (Look up Voltaire hearing sounds from putting an electrode in his mouth and one in his ear: the electrodes really haven't gotten much better.)

All smiles here

[MobileTatsu-NJG]

During the early tests of this device one technician was knocked unconscious after repeatedly urging her to shout "Hand, pick up the ball!"

Definitely Watch the Video

[IonOtter]

As it says in the article, it's totally worth seeing the look on her face.