Slashdot Mirror
MIT Microchip Could Someday Restore Vision
CWmike writes "Researchers at MIT have developed a microchip that could, one day, enable blind people to regain some level of vision. By combining wireless technology, eyeglasses equipped with a camera, and the chip, they should be able to restore at least some vision to people who suffer from retinitis pigmentosa or age-related macular degeneration, two of the leading causes of blindness, the scientists say. The chip, which is designed to be attached to the eyeball, would pick up images sent from the camera and electrically stimulate the nerve cells that normally carry visual input from the retina to the brain. The chip is sealed in a titanium case to keep water from leaking in and damaging its circuitry. At this point, the technology is not expected to restore normal vision, but MIT said it should provide the ability to navigate around a room or walk down a sidewalk. 'Anything that could help them see a little better and let them identify objects and move around a room would be an enormous help,' said Shawn Kelly, a researcher in MIT's Research Laboratory for Electronics. 'If they can recognize faces of people in a room, that brings them into the social environment as opposed to sitting there waiting for someone to talk to them.'"
Here are some questions I have about the chip:
- These chips/systems already exist. What's new about this MIT effort? The Computerworld article was very sparse.
- There's a great deal of bidirectional communication that goes on in normal eyes-- information not only flowing from eye to brain, but from brain to eye as well. As far as I know these tech just discards these signals. Is this important?
- Last I heard, this sort of technology was approaching 1000 effective pixels of visual information (assuming ideal electrode placement). Has this effort from MIT pushed this boundary? How does '1000 effective pixels' compare to the eye's effective resolution? Can we put normal vision in terms of pixel resolution?
- I've read about shunting tactile senses (for instance, the nerves on a person's tongue) over to a digital videocamera. I believe the military has done a fair bit of research into this. Could this sort of approach be viable for helping the blind function as well? Could it become the preferred approach since it seems less invasive than ocular- and neuro-surgery?
This is a total waste of time. Nature already has the best eye.. all we have to do is grow it.. But for the sake of technology, this stuff is great and can result in advances in other fields and products.. maybe someday allowing us to replace our eyes with totally hackable bionic eyes hooked into the internet and capable of playing all the porn we want anytime we want it.. shit, I missed my bus again..
Have you fscked your local propeller head today?
Gee, I wonder where they got this idea from...
You do not have a moral or legal right to do absolutely anything you want.
If this is as "great" as modern hearing aids, I will pass.
I have no desire to live with "coleco vision" for the remainder of my life.
I have to agree, even games that allow you to research multiple things at one time don't really give you any benefit to doing so. Space Empires, I'm looking at you.
Doing Tech 1 then 2 then 3 will get you the three techs just as fast as doing 1+2+3, with the added benefit that you'll get 1 and 2 a couple turns faster.
Personally, I'd do something like give a 1-10% advantage per additional tech you're researching to show the benefits of synergy and that a scientist suited for researching lasers is probably not going to do as good in biology.
Hmm... In a 'Research Point' setting, how about the points that go to a research field = RP/(number of research activities + 1).
1000 RP, 1 research topic (crash development) = 500 RP to that topic
1000 RP, 2 research topics (priority) = 333 RP to each topic, totaling 666
1000 RP, 9 topics = 100 RP each, each will take 5 times as long, but you'll get 9 techs in the time it'd take to get 5 serially.
Anyways, back more on the thread topic, growing new eyes would be great, but you still have the problem of restoring the nerve pathways and retraining the brain. It's entirely possible that you'd want a chip between the new eye and the brain to play translator, to make the transition earlier. Reducing therapy/usefulness from 5 years to 6 months (for example) might make it worth it.
I don't read AC A human right
... to keep a Cadillac driver on the road for a few more decades. Also, if they can put their glasses up on the dashboard, not being able to see over the steering wheel will no longer be an issue.
Have gnu, will travel.
Similar things have been reported for *at least* 30 years.
In the 1970's, I recall a sensor that clipped to eye glasses and connected to electrodes on the back of the user. I want to say that it was 16x16 or 32x32, but it provided enough "vision" to navigate and see objects.
A few months ago, iirc, was a report which used nerves on the tongue.
These reports are evolutionary, not revolutionary. A good thing, but it's not as if this is a breakthrough changing the world from "nothing to let the see" to "now they can see."
More efficient, easier to handle, lower cost--sure, but that's just the regular advancement of technology.
hawk
http://www.iblindness.org/
Just not profitable enough I suppose.
"Damn it! Where are my glasses!?"
Last I heard, this sort of technology was approaching 1000 effective pixels of visual information (assuming ideal electrode placement). Has this effort from MIT pushed this boundary? How does '1000 effective pixels' compare to the eye's effective resolution? Can we put normal vision in terms of pixel resolution?
To get an idea of the eye's effective resolution, think of a very high resolution monitor -- resolution so high that you couldn't tell the difference between it and a higher-resolution monitor sitting next to it. That's how much resolution the eye has.
Reviewers used to say that you didn't need a digital camera with a resolution greater than 5 megapixels, unless you were going to enlarge it. So 5 MP, which is about a 2,000x2,000 pixel line, sounds like a good guess. That's about 0.1 degree per pixel, which sounds about right.
That's not exactly right, because the eye doesn't take in a whole scene at once. It aims at different parts of the scene, with (in humans) the macula, which has higher resolution than the rest of the retina. So it's more efficient than a camera.
Besides resolution, contrast is also important. And the retina can recognize movement directly.
There's also a lot of image processing that goes on from the retina, to the optical nerve, through the processing centers in the brain. The retina figures out edges, points, movement, and color, and passes it on, then the brain figures out 3D stereoscopic information, objects, figures out what the objects are, recognizes them, and passes them on to other parts of the brain for higher processing. I think there are some error-correction circuits. There's a lot of great research on this.
This is oversimplified. There are people who enjoy correcting oversimplified explanations, and some of them know more about this than I do. I will leave it to them to enjoy themselves.
People with macular degeneration can already walk down a sidewalk or navigate around a room. They still have peripheral vision.
The macula is the spot at the center of the retina which has the highest concentration of receptors and the highest resolution.
It's damn useful, and it's hard (though not impossible) to read and identify faces without it. But macular degeneration spares the peripheral vision, so people can still get around. It's not "cane-tapping" blind. There are also some methods of using the peripheral vision to replace the macula with optical manipulation with fancy glasses, or digital manipulation.
Not that I would complain about the accuracy of a Slashdot story, but it's a teachable moment.
The real innovation of that MIT group is to place a titanium implant in the eye which can last a year in a pig. It's great work on the long, hard road to replacing the retina. I don't expect to see it in my lifetime, but hats off to anyone who can prove me wrong.
I'm sick and fucking tired of web sites that are a slim stip of content down the middle, with horseshit on the side.
"Second Sight expects the Argus II to be its first commercially available device, hitting the United States market in 2010. Mech said the company hasn't set a price, but suggested the Argus II would cost more than a cochlear implant. That could put the price between $60,000 and $100,000. Meanwhile, he hopes to start clinical trials in 2011 for a third generation of the device with more than 200 electrodes." [Science Notes 2009] http://scicom.ucsc.edu/SciNotes/0901/pages/vision/vision.html [Integrated Bioelectronics Research] http://ibr.soe.ucsc.edu/index.php?file=kop10.php