In 2002, six blind individuals arrived in Los Angeles from all across the country to undergo a groundbreaking surgery. They hoped for—but dared not expect—a miraculous result. They wanted to see again.

Dr. Mark Humayun, a professor of ophthalmology at the University of Southern California’s Doheny Eye Institute, had his fingers crossed too. He was the primary innovator of the tiny experimental device called a retinal prosthesis, and he was testing it for the first time. Humayun performed all six of the seven-hour implantation surgeries, in which each patient received an implant in one eye.

Humayun had spent 19 years designing a retinal prosthesis that would restore sight to the blind by electrically stimulating the retina. He wished his grandmother could have seen it. After all, she had been his guiding inspiration.

When Humayun was growing up outside Washington, D.C., he watched his grandmother, who practically raised him, go blind due to diabetic retinopathy, a form of retinal degeneration. As her vision faded away during his teenage years, he watched her zeal for life fade with it. When she went completely blind, Humayun said she essentially gave up and died shortly thereafter.

Humayun—from a family of physicians—initially leaned towards neurosurgery. But during medical school at Duke University, his grandmother’s struggle for sight lingered fresh in his mind. Then, during a clinical rotation in ophthalmology during his third year of medical school, he grew painfully aware of the millions who, like his grandmother, had lost their vision to retinal diseases.

“Her struggle charged me to stamp out blindness,” he said.

So he set out with a lofty goal for a 21-year-old: to build a device that would restore the circuitry of sight.

In normal vision, the lens of the eye focuses an image of the outside world onto a light-sensitive membrane at the back of the eye called the retina. Photoreceptors in the retina—rods and cones—absorb and code the constantly changing patterns of light into neuronal signals, or electrical messages for the brain.

Both retinitis pigmentosa (RP) and age-related macular degeneration (AMD) are diseases in which the light-sensing rods and cones of the retina degenerate and disappear, as happened to Humayun’s grandmother. As the photoreceptors die off over time, poor vision progresses to blindness. More than 25 million people, including 6 million in the United States alone, are blind or nearly blind from these diseases.

Humayun and his mentor Eugene de Juan, a professor of ophthalmology at Duke University, discovered that even though rods and cones stop working in a person with RP or AMD, the retinal neurons which communicate with the brain are often preserved to a great extent—even in blindness. They postulated that an implant might be able to bypass the ruined rods and cones and directly stimulate the related neurons.

“It was going to be incredibly difficult to engineer,” said Humayun.

He and de Juan quickly realized that one of the two of them would have to dramatically increase his technological expertise. Never one to turn down a challenge, Humayun went back to school, earning a PhD in biomedical engineering from the University of North Carolina at Chapel Hill while simultaneously enduring the grueling hours of a medical internship and residency. While other new doctors caught a few precious hours of sleep between shifts at the hospital, Humayun pored over books on microelectronic systems.

For someone so driven, the quietly charismatic Humayun has a mellow, observant demeanor. His youthfulness and humility belie his expertise. He never stops, according to his colleagues, but still he maintains a very natural calmness. He admits that he doesn’t have time to relax.

“I love what I do,” he said. “It is terribly addicting. That’s why I can do it almost non-stop.”

For his doctoral dissertation, Humayun created the blueprints for a device that would circumvent the damaged photoreceptors, deliver simplified electrical messages to the nerves, and let the brain sort out the rest. Then he, de Juan, and a team of doctors and engineers set out to build it. It would become the device implanted in those six lucky patients.

And it worked. The Model One retinal prosthesis restored both hope and sight where there had been neither.

To the untrained eye, a person with a retinal prosthesis just looks like a person wearing sunglasses—Oakley rip-offs to be exact. (Humayun said he was recently contacted by the president of Oakley about collaborating once the devices become more mainstream. “I think he just wants us to stop doing cheap knock-offs,” he said, laughing.)

A tiny camera is concealed either in the nose bridge or behind the tinted glass frame of the glasses and it videotapes the patient’s surroundings with the resolution of an average digital camera (4 megapixels). In real time, a microprocessor drastically simplifies whatever the camera is “seeing.” Neither color nor details remain. It also reduces normal vision’s 256 gray scales (levels of brightness) to 10.

The microprocessor converts the “dumbed-down” version of sight into electrical signals, which travel by radio wave from the processor to a receiver implanted in the skull. From the receiver, wires run internally to a 4x4 millimeter electrode array pinned to the retina with a tack the size of a human hair.

In the healthy eye, 1.2 million nerve fibers send information from the retina’s photoreceptors to the brain. The Model One used just 16 electrodes to electrically stimulate the retinal nerve fibers. That’s 16 electrodes for 1.2 million fibers. When they tested the very first patient, Humayun said he didn’t expect much. But remarkably, in what Humayun calls “a defining moment in his life,” the patients saw light.

And after a few sessions in the lab the patients could distinguish movement and shapes. Granted, a plate was described as a “saucer” of light and a knife a “runway” but the patients could decipher white shapes against a black background.

“We really thought they would just see black and white, maybe a little gray scale, maybe a little movement,” he said, with bewilderment in his voice to this day. “But there’s no way you should be able to use 16 pixels to differentiate a plate from a cup. That’s just staggering.”

When he speaks about the patients, Humayun’s passion shines through his cool facade. “I can’t convey the joy of the patients when they can see,” he said. “One hadn’t seen in 50 years.”

Over time, the Model One patients became increasingly confident in identifying the objects without having to touch them. “First it was a guess, then a pretty good guess, then it wasn’t a guess,” said Humayun, grinning. “They knew.”

Humayun, barely 45 years old, has accomplished more than most do in a lifetime. Highly regarded by his peers, his achievements include securing major research grants from the National Institutes of Health (NIH), National Science Foundation (NSF), and Defense Advanced Research Projects Agency (DARPA), as well as being the Director of a $34 million/10 year NSF Engineering Research Center on biomimetic microelectronic systems for neural prostheses. He has given more than 100 keynote and invited lectures around the world and also has a real interest in policy-making issues and raising public awareness, as if he has any free time. Already, he has served on the National Academy's panel to set policy in regards to science and engineering.

“Albeit you can have too much of a great thing, like candy or ice cream,” he said. “So I do take vacations with my family, skiing, etc.” He has two kids, Hannah, 9, and Lucas, 6. He said his wife Karen is very special and extremely tolerant. “None of this would be possible without her. But with the kids it’s different. I schedule time to do activities with them, and this is a schedule I do not break!”

Humayun is about to get even busier. In February 2007, he announced at the annual meeting of the American Association for the Advancement of Science that the FDA approved a clinical study to implant 50 to 75 blind people with the second generation of retinal implants. Named for the Greek mythological creature of 100 eyes, the Argus II, developed by USC and Second Sights Medical Products, is emblematic of the rapid learning curve of both Humayun and his team.

Whereas Model One took 17 years to develop, seven hours to implant, and had 16 electrodes, Model Two took just five years to develop, will take an hour and a half to implant, has 60 electrodes, and is a quarter the size of the first. It might be commercially available in as little as two years. The first generation retinal implant was made largely by hand—an extremely time-consuming and specialized task. Humayun said his team has now developed state-of-the-art microelectronic techniques that allow the devices to be easily scaled up (from 60 to 200 electrodes, for example) and to be produced en masse.

“Before we never would have thought to commercialize a 60-pixel device,” said Humayun. “But when we saw what we could do with 16…that’s a big change in plan and horizon.”

Humayun said the Model One implantations and results produced the most rewarding but most challenging times of his life.

“Model One was kind of like my first child,” he said. “I love her to death but it can be really hard work.” He sees Model Two like a second child: “Just as challenging and rewarding, and never a hint of being second in any way!”

The retinal prosthesis could ultimately restore sight to millions affected by macular degeneration or retinitis pigmentosa, such as Humayun’s grandmother. A Model Three that stimulates the retinal neurons using 1,000 electrodes, which may even enable reading and face recognition, is already cooking in the lab.

“It’s like cell phones,” he said. “They worked a long time ago, but now it’s, ‘Can we make them more functional?’ We’re taking it to the next level.”