Tuesday, May 23, 2006

Implantable Retinas Come Alive

An advance in triggering individual retinal cells could one day lead to implanted artificial retinas for treating blindness.

By Duncan Graham-Rowe

Encouraging neurons in the retina to grow dendrites into silicon-based eye implants could help make retinal prosthetics a reality. The dendrites grown in a controlled zigzag pattern (bottom) shown across an array of much larger vertical electrodes (top). (Courtesy of Neville Mehenti, Stanford University.)

The use of implanted retina chips to restore vision is still some way off. But scientists have taken a significant step toward that goal by finding a way to trigger the firing of individual retinal cells, by growing them on a silicon-based eye implant. The research represents a key step in making artificial vision considerably better and more natural than current methods used for prosthetic retinas.

One of the biggest challenges in developing artificial retinas has been finding a way to stimulate cells in the eye to fire individually rather than en masse. Now, by getting the retinal ganglion nerve cells to grow and plug themselves into a silicon device, it is possible, says Stacey Bent, a chemical engineer at Stanford University, whose research appears in the current issue of Biomedical Microdevices.

There are several different approaches used today in the attempt to develop retinal prosthetics. But the basic principle underlying all of them is the same: by stimulating cells within the retina, vision sensations can be elicited in the visual cortex. This is possible because for some common eye diseases, like retinitis pigmentosa and macular degeneration, only the light-sensitive photoreceptor cells in the retina are damaged. This means other types of cells in the retina and visual cortex in the brain remain intact and fully functional.

Until now, the method of choice for repairing these cells has consisted of using arrays of electrodes placed near the retina to stimulate the cells electrically. The trouble with this technique is that, apart from the electrodes being larger than the cells they're trying to stimulate, there is no way to isolate the electric fields in order to trigger individual neurons without triggering their neighbors.

Encouraging the cells to grow tentacle-like dendrites between the cell and an electrode, the strategy used by Bent, gets around this problem by creating a communication channel that stimulates the cell without invading or disrupting the structure of the retina.

The real payoff with this method, though, is the ability to make use of the preprocessing of the retina, says Bent. Until now, most research has focused on stimulating the retinal ganglion cells, the large cells that feed signals directly into the optic nerve. But this bypasses all the motion-detection and edge-detection processing carried out in the retina itself by a network of neurons called bipolar cells.

The ability to stimulate individual cells opens the door to the possibility of stimulating bipolar cells instead of ganglion cells, says Bent. "We'd be taking advantage of the existing circuitry," she says.