Counting Down to the End of Blindness

Knocking down children’s sandcastles isn’t everybody’s idea of fun. It certainly isn’t for a blind woman in Germany who risks destroying them each time she goes to the beach. Her walking cane topples them over.

That’s one downside to current assistive devices for the visually impaired, as heard by Dutch neural engineer Peter Meijer.

Not unlike other scientists around the globe, Meijer is working with the brain’s capacity to restore sight in people with complete blindness. As they reveal how neural pathways can be adapted and manipulated, they’re stoking hope for a system that may one day recreate normal vision — and far surpass technology available today.

The vOICe

How Meijer gets blind people to “see” is changing the way we think about the brain. In particular, his invention reveals the visual cortex’s ability to process spatial data from sound.

Spectrogram of Peter Meijer created by The Voice from a one-second soundscape (left). Peter Meijer shown in normal photograph (right).

Since 1983, Meijer’s been refining a system that enables people to view their visual environments through audio frequencies.

“Seeing with sound,” as he describes it, works thanks to his free software called The vOICe — the capital letters standing for “Oh, I see!” He says it’s been downloaded about 260,000 times on Android and hundreds of thousands of times on Windows.

To run the program users must also supply their own equipment, including a camera, smartphone or computer, and headphones. He recommends wide-angle camera glasses giving 120 and 180 degrees as a field of view.

“It’s a very different approach from many other systems,” Meijer says, referring to existing vision technologies on the market and in development. And he admits its concept can be hard to grasp.

The vOICe scans images taken from camera glasses, and then converts them into a series of noises. To the uninitiated, the sonic overlay sounds like it could be straight out of an episode of The X-Files.

“Learning to see with sound is not easy, it’s hard. It’s really hard” Meijer says.

He adds that most blind users need three months to get comfortable interpreting the soundscapes. Even then, the training isn’t over.

“It’s like with learning a foreign language. You can’t speak a foreign language fluently after three months, unless you are extremely talented,” he says.

The effort, however, may be worth it.

With practice people who are totally blind can gain up to 20/250 vision with The vOICe, according to research from the Hebrew University of Jerusalem and the University of Bath, England.

“It’s better than what the World Health Organization uses as its criterion for total blindness, and it’s also a lot better than what retinal implants offer you today,” Meijer says.

Neural coding and retinal prosthetics

Uncovering the brain’s wiring also holds promise for retinal implants.

In 2013, the U.S. Food and Drug Administration approved the first retinal implant for adults with retinitis pigmentosa — a degenerative disease that destroys photoreceptors and damages the retina.

Developed by Second Sight Medical Products, the Argus II Retinal Prosthesis System records images with a camera mounted on a patient’s glasses. These images are turned into electrical signals via an external video processing unit. They are then sent wirelessly to an array of 60 electrodes implanted on the retina.

The electrodes emit these signals as electrical impulses to the remaining retinal cells, allowing the person to see bright shapes. The company claims a maximum possible field of view of 20 degrees, which is comparable to a 30 cm ruler held at arm’s length.

Users boast about how retinal implants let them visibly notice objects, and teams in the United States and Australia are developing their own prosthetics.

But still, the technology in general faces criticisms for its narrow visual field and limited image quality

Sheila Nirenberg, an American neuroscientist at the University of Cornell, has won a “Genius Grant” from the MacArthur Foundation for her work to improve the latter.

According to Nirenberg, who has written a paper and given several media interviews about her work, current retinal prosthetics let people see bright lights and high-contrast edges. Yet it’s nothing close to 20/20 vision.

While the retina’s output cells do fire, they fail to do so in the right pattern. The result is an unclear image, she says.

So instead of trying to improve resolution to fix the issue, as scientists have attempted with limited success, Nirenberg has taken a different approach.

She’s uncovered the retina’s neural code. And by programming it into an injectable chip, she claims she can make a blind retina fire the same way a normal retina would. The outcome is a much more recognizable image.

The Brain Implant

Fifteen years ago, the big news in visual prosthetics was William Dobelle’s artificial eye. It set the record as the first to successfully give back vision to a blind person.

Though revolutionary, the system was clunky and controversial.

Diagram of William Dobelle’s artificial vision system.

Sixty-eight electrodes were implanted on the surface of Jeremiah Teehan’s visual cortex. And two holes were drilled into the man’s skull, through which wires connected his implants to a computer and signal processor. A small camera on his glasses sent images for processing through this network up to his brain.

After living 36 years in blindness caused by blow to the head, Teehan could see once more — albeit what he saw were simple bright dots outlining objects in his environment.

Following the procedure, the U.S. Food and Drug Administration banned the testing of visual neuroprosthetics in humans. Dobelle then moved his headquarters from New York to Lisbon, Portugal, to keep his research going. He died in 2004 due complications from diabetes.

Today, no visual brain implant is on the market for humans, and testing is limited to animals.

For the past 12 years, Mohamad Sawan, an electrical engineer from L’École Polytechnique in Montreal, Canada, has been trying to change this.

“This research, as you can imagine, is longer than anyone can expect. It’s not really a project as a career,” Sawan says.

In fact, Dobelle began developing his system in 1968. And scientists at the Illinois Institute of Technology in the United States have been working on their own prosthetic since 2000.

Sawan’s method differs from Dobelle’s as his implant is placed inside the brain and is wireless. He’s tested the system on rats and monkeys, charting patterns received by the primate’s optic nerves

Diagram of Mohamad Sawan’s intracortical visual prosthesis.

According to Sawan, his visual prosthetic is safer and more precise as it rests under the brain’s dura. Surface implants require a stronger current, upping the risk of seizure, electrode corrosion and tissue damage.

The technique is also hailed as more likely to give better visual acuity and field of view than retinal implants.

“When you stimulate a different end, you are sending information, we don’t know how it’s filtered and arrives to the brain,” Sawan says.

“But when you are at the brain level, you are mapping all the area as you exactly you want those pixels.”

Sawan is yet to begin testing in humans. Doing so would be costly, and his team needs to be sure there is no potential for harm. It’s hard to give an estimated date, he says, because the deadline keeps getting pushed back.

“We were positive a few years ago, we said maybe 10 to 15 years, from year 2000, but still we are not sure,” Sawan says.

“Vision in a human is complex. It’s different than of course cochlear implant and pacemaker… because vision is somehow more concentrated inside the brain and it’s not really fully understood.”

Image sources: Second Sight Medical Products, POLYSTIM Neurotechnologies, The Dobelle Institute

This article was originally published as part of the “Health Technology and the Future” series on Beacon Reader.

Originally published at TechAble World.