MIT and Harvard Medical School collaborators are producing a sophisticated engineering tool that electrically stimulates the retina to provide vision of a sort for people who are totally blind. The question is: will it be useful vision?
Their goal has been to develop a silicon chip eye implant that can restore vision in patients with retinitis pigmentosa, the leading cause of inherited blindness, and with age-related macular degeneration, the leading cause of blindness in the general population.
The Retinal Prosthesis Group, an academic consortium led by Joseph F. Rizzo, a neuro-ophthalmologist and research affiliate with MIT's Research Laboratory for Electronics, and John L. Wyatt, professor of electrical engineering and computer science, has been working for about 15 years on how to bypass a diseased or damaged retina to get images to the brain. The group consists of 20 researchers, among them materials scientists, software engineers, ophthalmologists and retinal physiologists.
"See" is a relative term. The researchers do not at the moment think the goal is to have people with virtually no usable retina suddenly able to read The New York Times. The idea is that with a retinal prosthesis, a person may be able to find a door in a room or walk down the street without the aid of a cane, or recognize a face or whether the face they see is smiling or frowning.
These are major quality-of-life impediments that may be improved if their work is successful, the researchers say. In the meantime, they face substantial obstacles in wedding technology and biology to create a safe, workable device. Among other challenges, the largely transparent retina is a quarter of a millimeter thick and the consistency of wet tissue paper.
Rizzo, director of the Boston Veterans Administration Medical Center and a neuro-opthalmologist at Massachusetts Eye and Ear Infirmary, recently described the results of clinical trials of a prototype retinal prosthesis to MIT brain and cognitive sciences students and faculty. Under local anesthesia, five blind subjects, plus one person with a normal retina who was going to lose her eye because of a surrounding cancer, were fitted with an ultra-thin electrode grid that was introduced through their dilated pupils and temporarily attached to the retinal surface. The grid bypasses the individual's defective rods and cones by stimulating healthy ganglion cells with a tiny electrical current.
The patients described what they perceived when their retinas were electrically stimulated in different ways. The five blind patients, some of whom had been blind for decades, also drew what they saw. One woman said she saw clouds. Half the time, Rizzo said, there was a correlation between the pattern of stimulation and what the patients described. A single stimulus could result in a cluster of small precepts, hence the clouds. But half the time, the people saw things that didn't make sense to the researchers.
There are plenty of challenges associated with integrating electronic devices into the body, especially in a sensitive area such as the back of the eye, just millimeters from the optic nerve leading to the brain.
For instance, how much electricity does it make to stimulate a blind retina? No one knows how much current the retina can bear, especially over a long period of time, so they use electrodes ranging from 100 to 400 microns in diameter (the diameter of a human hair is 50-100 microns).
"We still don't know if you could drive the retina for 20 years without killing it," Rizzo said. In addition, heat is given off during energy transfer, but luckily the vitreous fluid in the middle of the eye dissipates it. This same salty vitreous liquid would require any wires and connections to be made of inert and durable materials.
The organization of the retinal nerve cells is precise, designed to send light in patterns to specific cells designed to receive it, but unlike light, electric stimulation hits all the cells equally. Rizzo says they are exploring ways to narrow the electrical field to hit more specifically on the retina. In patients with retinitis pigmentosa, there is a loss of photoreceptors, but the researchers work with the relatively large number of retinal ganglion cells that survive.
In the clinical trials, stimulating one electrode created an image akin to a circle, while stimulating two adjacent electrodes created a line or long thin rectangle. If the stimulus could reliably form perceptions of certain shapes, it might be possible to use the system to create letters, numbers and maps.
The researchers were frustrated but not surprised to find that stimuli in the shape of letters did not yield recognition. "What we're trying to do is merge traditional studies on the retina's physiology with human perceptual results," Rizzo said. The researchers acknowledge that they are working on a Herculean task and have a long way to go and many questions to answer before they succeed.
A version of this article appeared in MIT Tech Talk on December 3, 2003.