Neil Osterweil

May 02, 2011

May 2, 2011 (Fort Lauderdale, Florida) — Seeing with the tongue instead of the eyes sounds like something an extraterrestrial on Star Trek might do. But a novel sensory apparatus that transmits environmental cues to the brain via a sensor array placed on the tongue could be a low-cost and noninvasive alternative to retinal implants for the blind.

In a study involving 21 blind and 6 normally sighted but blindfolded control subjects, participants using a spectacle-mounted video camera coupled to a sensory array held in their mouths were able to identify about 40% to 60% of life-sized obstacles in their path over a 40-foot course, and to improve their walking speed to nearly that of walking an unobstructed course, reported investigators here at the Association for Research in Vision and Ophthalmology (ARVO) 2011 Annual Meeting.

"Electrical stimulation of the tongue via signals sent from a spectacle-mounted video camera results in a sensation akin to vision, and enables a perception of one's surroundings that can be useful. The extent of the visual field is sufficiently large to enable navigation with obstacle avoidance," Thomas Friberg, MD, professor of ophthalmology at the University of Pittsburgh, Pennsylvania, and colleagues report in a poster presentation.

Although the device he and his colleagues studied is a prototype and not quite ready for commercial development, it is a deceptively simple and promising approach to a complex problem, said Dr. Friberg in an interview with Medscape Medical News.

"As retinal specialists, we have lots of patients who come to us who can't see, and to try to rehabilitate them is an important part of our job, rather than just throwing up our hands and saying there's nothing we can do," he said in an interview with Medscape Medical News.

One Camera + One Tongue = Two Eyes?

Implantable retinal chips are invasive, expensive, and require exhaustive testing and investigational review board consent. In contrast, the device, dubbed BrainPort (Wicab, Inc, Middleton, Wisconsin), relies on a video camera coupled to a 400-electrode sensor array placed on one of the most sensitive parts of the body — the tongue. The system produces electro-tactile feedback that allows users to "see" or interpret objects in their ambient surroundings, Dr. Friberg noted.

"It's very small, so you walk down the street with what looks like dark sunglasses and a tiny wire in your mouth," he noted.

With a few hours of training, patients can be taught to use the system to sense and avoid objects in front of them. Over a longer training period, many patients can begin to identify shapes, saying, for example, "that looks like a box," or "that looks like a chair."

Whether patients who use it are actually "seeing" is currently anyone's guess, Dr. Friberg acknowledged.

A retinal specialist who was not involved in the study told Medscape Medical News that he is excited to see such a device on the near horizon.

"As a clinician, I've got a number of patients who are always asking me: 'surely there's something in the world of research that can help me.' And I've had to say 'sorry, there's nothing that can really help you.' Then you see something like this that has suddenly popped up. I've been to ARVO for years and years and years and I don't think I've seen anything like this before," said David Broadway, MD, consultant ophthalmic surgeon and honorary senior lecturer at Norfolk and Norwich University Hospital in the United Kingdom.

"The only thing that is weird about it is that you can hardly believe it can work," he added.

Study Design

Dr. Friberg and colleagues enrolled 21 blind patients with only light perception vision or worse, and 6 sighted patients as control subjects. Participants received about 20 hours of training on the current prototype of the device, which consists of a camera with a 73 degree field of view, coupled to a 1-inch square array containing 400 electrodes.

The participants walked a 7 × 40 foot obstacle course with 3 levels of difficulty, and with each level having 3 iterations under varying conditions of illumination.

The primary outcome was the percentage of preferred walking speed achieved, measured as the ratio of time to walk 40 feet with no obstructions, divided by the time to walk the obstacle course with the device.

At first, preferred walking speed with the device slowed significantly from baseline (=.001) because patients were searching for objects with the device. But as they gained experience, their walking speed increased, approaching that of the time to navigate the unobstructed course.

"As transducers become sophisticated with denser electrode arrays, the utility of this device will likely be enhanced," the investigators write.

The study was supported by grants from Research to Prevent Blindness and the National Institutes of Health. Dr. Friberg and Dr. Broadway have disclosed no relevant financial relationships.

Association for Research in Vision and Ophthalmology (ARVO) 2011 Annual Meeting: Poster abstract 109/A244. Presented May 1, 2011.