The Circuitry of Sight

Prosthetic Networks Could Power Vision

By Todd Dills


sp2010_circuitry
Franklin Amthor's microelectrode arrays stimulate the specialized nerve cells that transmit visual information to the brain.

Trained as an electrical engineer, UAB psychology professor and neurophysiologist Franklin Amthor, Ph.D., applies his knowledge to the mother of all motherboards: the human brain. His research into the complex connections among neurons has led to a potential solution for progressive eye diseases that cause blindness: tiny electronic networks called microelectrode arrays that could do the job of damaged neurons and help restore sight.

The arrays are designed to stimulate—or record the stimulation of—retinal ganglion cells. These specialized nerve cells transmit visual information from the outside world to the brain. The long-term goal, Amthor says, is to provide a visual prosthesis for patients with diseases such as retinitis pigmentosa, a condition in which visual information doesn’t reach the brain, resulting in tunnel vision and progressive blindness. While these diseases affect the retina’s photoreceptors, which collect visual stimuli, they leave the ganglion cells intact. Amthor’s technology, coupled with tiny cameras, may be able to mimic the actions of biological photoreceptors and restore the link to the brain.

Amthor continues to refine his design to help his invention realize its full potential. A major challenge, he says, is figuring out how to make his device small and streamlined enough to fit on the retina. The arrays would be about a square millimeter in size, Amthor explains. “They have to be very tiny so that when you move your eye around, the inertial load of the array wouldn’t damage the retina.” He also must address the toothlike projections in his current model, which he calls “a serious manufacturing problem.” Amthor has made strides from his previous prototypes, however, using carbon fibers instead of silicon-derived elements to conduct the nerve impulses. One reason he chose carbon fibers is because of their biocompatibility.

Amthor began studying real neurons after investigating artificial ones. As an undergraduate in bioelectronic engineering at Cornell University in the late 1960s and early ’70s, he studied under Frank Rosenblatt and Henry Bloch, primary developers of the Cornell Perceptron. This machine, Amthor says, was one of the first “artificial neural nets”—essentially acting like a human brain by utilizing electronic equivalents of neurons to enable certain kinds of learning.

Captivated by the idea “that you could model the brain with a computer,” Amthor spent hours attempting to teach the perceptron to read text. Along the way, he became an expert on artificial intelligence—and began investigating its potential applications to solve real, biological problems.

Though Amthor’s research has been supported by National Eye Institute grants in the past, he continues to develop his innovative prosthetic microelectrode arrays even when funds are scarce. He says he is glad to be part of Birmingham’s vision-research community, which he places among the nation’s top 10 in part because of its collaborative potential, and he remains adept at making do with what’s available. “With my engineering background, I’ve been good at building my own equipment and been able to keep going when a lot of people might not,” Amthor says.

 

More Information

UAB Department of Psychology

UAB Vision Science Research Center