Eyes that Should Be Blind

PRINT     E-MAIL STORY    SHARE

FACEBOOK

REDDIT

NEWSVINE

DIGG

DEL.ICIO.US

BLINKLIST

Thursday, June 19, 2008 | It's an advance that's both astounding and not very useful.

With the help of a high-tech implant, blind people can see bright flashes of light -- and nothing more.

Salk Institute researcher E.J. Chichilnisky describes his studies of the retina. Photo: Sam Hodgson

There are 1.25 million obstacles to bringing more sight to the sightless. That's the number of nerve cells that transfer electrical signals from the eye's retina to the brain, which transforms them into images.

No one knows how they all work. The retina, which acts as a kind of transfer station between the outside world and the brain, remains a mystery. 

But in a laboratory at the Salk Institute in San Diego, the retinas of animals that have just died still see -- and that is opening a window into the inner workings of the sense of sight. Sitting in a multi-million-dollar machine, the retinas gaze at images on a monitor and transmit electrical signals to brains that aren't there anymore. Instead, a computer is on hand to capture the messages and preserve them for scientists to pore over.

Neuroscientist E.J. Chichilnisky is in charge of the project and gaining new insight into the roughly 20 types of nerve cells that make up the 1.25 million.

Related Links After Building Careers Here, Chinese Researchers Drawn Home (April 17, 2008) How San Diego Researchers Walk Through the 'Valley of Death' (April 9, 2008) New to Local Political Game, Bio and Tech Score Early Victory (March 27, 2008)

"We're trying to figure out how information is analyzed and transmitted by the retina in order to understand what those signals are. That's a very big job," said the pony-tailed Chichilnisky, a slender middle-aged scientist who works in a small office at the Salk Institute.

It's such a big job that donors -- including the federal government -- have spent millions on Chichilnisky's research over the past 10 years.

Chichilnisky estimates that hundreds of scientists around the world are studying the retina, which sits at the back of the eyeball and is often called the equivalent of a camera's film because images are imprinted on it.

Chichilnisky's area of research, however, is unique.

Scientists at the Salk Institute track the electrical signals that are sent out by animal retinas that gaze at images like this one. Photo: Sam Hodgson

He and his colleagues are tracking the electrical signals that the retina sends to the brain through so-called retinal ganglion cells. (Since they're part of the brain, they're also known as neurons or nerve cells; they send signals through a filament called an axon or nerve fiber.)

The brain is in charge of creating images of the world from the information sent in by the eye. According to scientists, the retina serves as a translator, "seeing" through the eye's lens and converting the visual world into a language the brain can understand.

No one can speak that language, at least not yet.

"A big question is trying to understand how all of that information is encoded in those (electrical) spikes," said Marla Feller, a professor of neurobiology at the University of California, Berkeley.

In order to study retinas, Chichilnisky must first get them. Most of the animals in his studies are rats that other scientists pass along to him after euthanizing them. Rarely, just a few times a year at most, the laboratory gains access to the retinas of deceased primates, which are most similar to those in humans.

To continue to function, retinas must be removed within four minutes of death and preserved in a solution of salt and sugars, Chichilnisky said. They will continue to process images and send electrical signals for up to 24 hours.

Other researchers use electrodes to monitor individual neurons and track the electrical signals they each send to the brain. Chichilnisky's laboratory takes a different approach, monitoring the activity of hundreds of neurons at once.

The lab relies on a machine that uses tiny electrodes that are a diameter of less than five microns across -- about 5 percent of the width of a human hair.

"We started from nothing," said the machine's developer, University of California, Santa Cruz physics professor Alan Litke. "No one had ever done anything like this before. We had to develop everything from scratch, developing new kinds of integrated circuits, electrode arrays, and completely new software."

Chichilnisky said one of his current focuses is understanding the different types of retinal ganglion cells and their jobs.

It appears that they transmit different types of information about images to the brain, tracking things like movement and lightness or darkness; some may transmit aspects of an image in higher or lower resolution.

In the Salk laboratory, the retinas of experimental animals are placed flat onto a tiny array of electrodes in a machine that looks a bit like a microscope. The still-functional retinas look through a lens -- just as in the eye -- and see a computer monitor that displays various patterns, including one that looks like television static.

A computer tracks the electrical activity of the retina and converts it into a terabyte of data, Chichilnisky said. (A terabyte, according to one calculation, could hold 1,000 copies of the Encyclopedia Britannica. A 100GB hard drive is a 10th that size.)

Chichilnisky said his research is largely a matter of curiosity, simply gaining a better understanding of how the brain works. But he's also trying to help the blind. An estimated 5 million to 10 million Americans suffer from partial or total blindness caused by diseases like inherited retina pigmentosa and age-related macular degeneration. In many cases, retinal cells die and cannot be replaced.

But the rest of the brain apparatus that allows us to see may still work perfectly, meaning that a prosthesis could -- theoretically -- replace the retina and restore sight.

To that end, Chichilnisky is working with a company called Second Sight that is testing early versions of a retinal implant on volunteers.

With the help of a tiny camera mounted in eyeglasses, the device transmits image information through 16 electrodes, the equivalent of neurons. So far, Chichilnisky said, the volunteers only report seeing flashes of light, no surprise since the device sends information through 16 inputs compared to the eye's 1.25 million.

"It doesn't help them to do anything right now -- cook their meals, recognize their kids, or read. Their vision has not been restored," he said.

But eventually, the 16 inputs should grow to hundreds, he said. "We want to produce something like a high-resolution camera that modifies and encodes the visual image in a way that interfaces correctly with the brain. We want to be able to produce an image that's fine-grained enough that you can read, that has the nuances of shades and colors, that can identify a face and sense that something is moving, a car is coming your way," Chichilnisky said.

He thinks it will be years before the retinal implant is perfected, and even longer -- lifetimes, in fact -- before scientists understand the how the brain works. But he's not discouraged.

"The brain is the most interesting puzzle of all," he said. "It's the greatest frontier in science."

Randy Dotinga is a San Diego based freelance writer. You can contact him at rdotinga@aol.com. Or set the tone of the debate with a letter to the editor.