THE VISUAL PROSTHESIS envisioned by John Pezaris would provide the patient with a special set of glasses that has a small digital camera mounted in the lens. The camera's wire communicates to an external signal processor that translates the image from the camera into neural impulses transmitted to an implanted stimulator, which delivers images to the visual system. Image: Courtesy of J. S. Pezaris, adapted with permission from D. H. Hubel
The ability to see requires healthy eyes, but it also requires that signals can get from the eyes to the parts of the brain involved in vision. A Boston neuroscientist hopes to deliver a ray of hope to the blind by bypassing eyes and optic nerves damaged by illness or head trauma and sending image information directly to the regions of the brain that process them.
The prosthesis proposed by John Pezaris, an assistant in neuroscience at Massachusetts General Hospital (MGH) in Boston—at least as it's envisioned at this early stage—would be worn like a pair of eyeglasses, with digital cameras over a person's eyes that connect to an array of electrodes implanted in the brain. Although this doesn't promise to restore normal vision, "a remarkable amount of information can be conveyed in a relatively small number of pixels," he says, that would allow people to perhaps identify simple objects and even recognize faces.
The technology still has a lot of obstacles to overcome—the need for digital imaging that can adequately substitute for normal vision and the risk of infection resulting from brain surgery, to name two—but success could have a life-altering impact on the tens of millions of people worldwide suffering from impaired vision.
It's difficult to say exactly what a person wearing the prosthesis would see, but most likely it would be like looking at pieces of a puzzle. Still, as the ability to deliver higher resolution images to the brain improves, so would the effectiveness of the prosthesis. "We've been working on trying to understand more about the percepts, or visual events, we're creating," Pezaris says, including what images look like, how big they are, how long they appear and whether they are in color or black and white. "We don't have a really solid handle on any of this now."
Such a prosthesis could be useful to many people: The World Health Organization estimates there are 45 million people who are blind, and they expect this number to double by 2020. "You can't underestimate the scope of the problem, particularly as the population gets older," says James Morrison, a physiologist and principal investigator in the Retinal Prosthesis Group at the University of Glasgow's Institute of Biomedical and Life Sciences. The numbers may get worse as the population ages: Morrison notes that 8 percent of U.K. residents over the age of 65 suffer from age-related macular degeneration.
Determining the number of people who stand to benefit from the technology is more difficult because different people suffer from different degrees of blindness, but it's fair to say that candidates for Pezaris's implants are more likely to be those who are totally blind, whether from birth or due to a traumatic brain injury. Pezaris says he would like to eventually help soldiers suffering from blindness due to blast damage, for example.
People who have been diagnosed as "legally blind" but who have some residual sight—such as New York State's new governor, David Paterson—would not be good candidates for the surgery. In these cases, Pezaris says, "the benefits of the device do not outweigh the risks of the surgery."
There are several challenges to the development of visual prosthetics. One is the inability to replace the function of cells in the retina with a digital camera. "The retina is a very complex device, enabling our visual system to adjust to tremendous changes in light intensity," says Thomas Serre, a neuroscientist at M.I.T.'s McGovern Institute for Brain Research's Center for Biological and Computational Learning. "It's not clear how far we could go just by acquiring an image with a video camera." Another issue is the likelihood that images resulting from brain stimulation would appear distorted when compared with the actual image. "Microstimulations are hard to do, and sometimes some weird effects happen," he adds, referring to a technique that activates a cluster of nerve cells by zapping them with a weak electric current.
Much of Pezaris's work is based on "best guesses," he acknowledges. He and former Harvard Medical School colleague Clay Reid first described their research into visual prostheses in a paper published last May in the Proceedings of the National Academy of Sciences USA. The paper described experiments performed on two adult macaques with electrode brain implants to determine whether the primates would react when electric stimulation was applied. The researchers were able to show that microstimulation in certain areas of the brain does create a percept that the brain interprets as optical input, or something that can be "seen." The researchers were unable, however, to determine some of the details of the impression, such as its exact size or shape, although they knew that the percepts were, "like points of light, or single pixels," he says.
"On the scientific side, this is a tremendous achievement," Serre says of Pezaris's work, adding that it's one of several important advances in visual prosthetics over the past five years. Whereas some of the other types of visual prostheses propose stimulating different areas of the brain from the one Pezaris is targeting, another promising project known as the Argus II Retinal Prosthesis System is expanding its U.S. clinical trial into Europe. The project, headed by Mark Humayun, a professor of ophthalmology and biomedical engineering at the Keck School of Medicine of the University of Southern California, involves an implant consisting of 60 electrodes attached to the retina that conduct information from an external camera to the retina to provide a rudimentary form of sight to patients with the implants.
Pezaris has also enlisted the help of Emad Eskandar, a neurosurgeon at MGH who specializes in deep-brain stimulation, which has been used to treat Parkinson's disease and monitor neural activity in people suffering from seizures. Deep-brain stimulation involves the surgical placement of electrodes in the brain to deliver stimulation to targeted areas that control movement, similar to the way pacemakers are used to maintain a healthy heart rate.
The best way to test the effectiveness of Pezaris's prosthesis is to implant the electrodes in a human patient, a step that requires funding, approval from MGH's institutional review board and a volunteer whose condition is amenable to the treatment. Pezaris, who joined the hospital in November from Harvard Medical School's Department of Neurobiology, says he is in the process of trying to meet all of these criteria but does not know when a human trial might happen.
Pezaris is hoping to have a functioning device ready for human testing within a few years, noting that his approach allows him to leverage a lot of existing medical technology. Yet he's careful when describing his project's prognosis: "since it's so early on in the process, and we don't want to overpromise or take unnecessary risks."