Artificial Vision

By Amy S. Hansen

The futuristic device might work like this. The video camera takes a picture of whatever is in front of it, then sends the picture to the silicon imaging chip. This chip changes the picture into a simple pattern of black and white spaces. Using radio waves, the chip sends the image to the eye. On the back of the eye, another electronic chip receives the picture and “shows” it to nerve cells called ganglia. Those cells pass the information to the brain, which can then “see” the picture.

In science fiction, a character may go blind but then use a pair of ultra-high-tech artificial eyes to see again. In laboratories, re-searchers are working to make that fiction a reality.

Here’s the plan: A blind patient will have a tiny device put into one eye and will wear glasses that hold small machines. And with this system, that person will read a magazine, watch a movie, and see old friends again.

Working together, surgeons in California and engineers in North Carolina have already built early versions of the system’s parts. They expect that someday their devices will be able to replace a part of the eye called the retina.

What Is a Retina?
The retina is a wet membrane in the back of the eye that holds a layer of cells called cones and rods. These cells play a key part in vision. They detect any light that comes into the eye. Then they send signals to other cells, giving them information about that light.

That information is passed to still another group of cells, called the ganglia, which are also in the retina. Their job is to send the information to the brain through the optic nerve at the very back of the eye. A person sees what’s in front of his or her eyes when the brain uses all that information to create pictures.

Injuries or diseases can make a person blind by damaging any part of this system. Some diseases hurt the retinas, destroying the cones and rods. Worldwide, more than 10 million people have lost their sight, mostly in old age, because of these kinds of diseases.

This group of surgeons and engineers is developing an artificial retina for people whose rods and cones have been destroyed. The light-sensing cells are gone. But the diseases have not harmed the ganglia or optic nerves.

The scientists wondered: Could a device take the place of the rods and cones? Could a tiny machine give some simple pictures to the brain by stimulating the ganglia?

Here’s how two familiar sights would look to a person using the artificial-vision device. The pictures on the right show the same ball and the same face after the photographs have been changed into simple patterns of black and white spaces in a 32-by-32 array. The objects will be easier to recognize when they move and when the user can also hear or touch them. By looking at the pictures on the left through a magnifying lens, you can see that they are also made up of dots. More (and smaller) dots make a sharper picture.

To find out, the team experimented with fifteen volunteers. The volunteers were blind when they walked into the surgical suite. They were still blind when they walked out again. But while they were there, fourteen of them could see again—if only a little.

The process wasn’t magic, but it was startling. Surgeons inserted a tiny electrical probe into the back of each volunteer’s eye and then moved the probe around. A small electrical signal passed from the probe to the retina.

The blind volunteers saw spots of light plus shapes that the doctors made by moving the probe. Now the scientists knew that electric signals could give information to the visual system, even without rods or cones.

The team put two of the volunteers through a more complicated test. The idea was to create a pattern on the retina in much the same way a TV or computer creates a picture on a screen—with a pattern of dots.

Using twenty-five tiny probes arranged on a square holder, the doctors created simple patterns of electric signals on the retinas.

One volunteer correctly identified a square. The second volunteer recognized an H, the first letter of his name. The grid of probes can’t be used outside the laboratory yet, but it’s a start.

An Artificial Retina
The top surgeon on the team is Dr. Mark Humayun at the University of Southern California. He is learning what happens when a retina is damaged. The engineers on the team are led by Dr. Wentai Liu at North Carolina State University. They are inventing devices that will recreate sight.

In the future, the device will work something like this: When the high-tech glasses are switched on, a tiny video camera in the glasses takes pictures of the outside world. A radio signal sends power and information from the camera to an electronic chip, like a computer chip, that has been implanted in the eye. Then the chip uses many tiny electric probes to stimulate ganglia in different places, giving the retina patterns of dots. And those patterns are the pictures from the camera.

How well will people with artificial vision be able to see? When using dots to make pictures, the more dots you have in one space, the more details you can see. One of the first retina chips had only a 5-by-5 array of probes or dots. It was used in the experiment that let blind volunteers see a square shape and the letter H
.
A new chip will have a 32-by-32 array, which should give the user some more details, like the examples on this page.

Dr. Liu is planning a chip that will handle at least a 250-by-250-dot display. That should give the retina a sharp enough picture for people to read a magazine or newspaper.

The team expects to test the chip on more volunteers in a few years. And when they do, they’ll be moving artificial vision out of science fiction and into reality.