In the United States, anyone with a vision of 20/200 or less is considered legally blind. The number doesn't refer to the right and left eyes - it actually means that the person sees objects 20 feet away with the same sharpness that a normal person can see an object 200 feet away. That's why normal vision is referred to as 20/20.
For some time, doctors have experimented with biomedical implants that boost eyesight among the visually impaired, interacting directly with the neurons that send visual information to the brain. These haven't been super-successful, though - in about 86 percent of cases, there's been no improvement.
Now, however, a team of physicists from the University of Oregon believe that they might know why. They pin the problem on the electrodes that connect to the neurons - specifically, that they're the wrong shape. So they're designing new ones.
Traditional electrodes are based on traditional geometry, like squares. But the team's new electrodes instead mimic the design of the neurons they interact with - they're shaped like , with repeating patterns at many scales.
More neurons, less voltage
In simulations, the team showed that not only was the fractal design able to stimulate 90 percent more neurons in the retina, it did so while using less voltage than a traditional implant. "A fractal implant should allow us, in principle, to deliver 20/80 vision," said Richard P. Taylor, who leads the institute where the research was conducted.
The result is an implant that can efficiently utilise more pixels in the crowded space at the back of the eye.
It's still early days, so human trials are some way off. The implants are too large right now, so the team is working on designing miniature versions that can be tested in mice. Eventually, however, the discovery could bring major quality of life improvements to people suffering from retinal diseases.