Technology Restores Speech, Movement

Dec. 26, 2007 — -- The day may soon come when those who can't speak will be able to talk, those who cannot walk will be able to move freely, and those who cannot move at all will be able to turn on the television set and turn down the lights.

Scientists and engineers are making remarkable progress in developing the technology needed to intercept signals sent by the human brain and translate those signals into various types of actions. The possible applications are virtually limitless, with the potential to bring profound relief to millions of people who have lost some of their bodily functions because of disease or traumatic injuries.

The technology has advanced well beyond the surgically implanted sensors of just a couple of years ago. Today's sensors are noninvasive. They are featherweight, tiny, and need only be held against the skin.

Some of the firms that are bringing the technology to the marketplace are almost as new as the technology itself. The Ambient Corp. of Chicago is just slightly over two years old, but it expects to introduce a device called the Audeo by the middle of next year, restoring speech to some who have lost it.

The company was founded in October 2005 by a couple of whiz kids from the University of Illinois at Urbana-Champaign -- Michael Callahan and Thomas Coleman. The two have been working for several years to create a system so sensitive that it can read the signals sent by the brain to the vocal chords, and send those signals to a computer that can pronounce the words that the person can no longer say.

As a demonstration project, Callahan and Coleman built a wheelchair that can be controlled entirely by the operator simply thinking about saying which way to turn the wheelchair, and whether to move forward or backward. The Audeo intercepts the signals sent by the brain and turns them into commands to move the wheelchair.

Actual production of the wheelchair is several years away, Callahan said in an interview, but not because the system doesn't work. It's just that it doesn't work exactly right every time, thus raising serious concerns about liability.

"If either the person makes a mistake using the device, or the device itself makes a mistake that's tied to movement, that could be bad for everybody," Callahan said.

So the wheelchair is on the back burner, but a communications device is moving full-speed ahead.

The two entrepreneurial engineers have been working with the Rehabilitation Institute of Chicago, one of the most respected rehabilitation centers in the world, to help people "speak" after losing their ability to talk because of disease or injury.

"Patients who have come from all over the world get to be involved in some of the testing and development work," Callahan said. The institute is aggressively pursuing the technology.

All this began years ago, when Callahan and Coleman were students in the engineering department at the University of Illinois. The department is famous for its emphasis on getting students to teach themselves and then move into the marketplace to solve global problems.

An early project by the two students involved creating a computer mouse that could be controlled remotely by "brain waves," a hot issue in many labs at the time. Neither student was a computer guru, so each relied on an innovative bit of software from National Instruments in Austin, Texas. The software, called LabVIEW, was a nice fit with their training at the university in that it encourages personal initiative in computer programming.

"It's extremely empowering," Callahan said.

The importance of that early research is that it launched the two on a determined course to help the human brain carry out its commands in patients who can no longer do it themselves.

The wheelchair, created partly as a public relations tool so people could better understand the technology, is surprisingly simple in its operation. A tiny sensor, mounted in a collar, is positioned directly over the vocal chords, where it can "read" the electronic signals sent by the brain.

"We pick up a very small instruction signal on the surface of the skin because the material underneath the skin, the blood and the tissues, are moderately conductive," Callahan said. "We can pick up the signal and turn it into useful things," in this case, speech.

In order to work, the user has to do more than just think about what the patient wants to say, or what the wheelchair is supposed to do. The patient must think about saying what he or she wants to do, thus sending the signal to the vocal chords.

"If you are just thinking something, like turning the wheelchair to the right, our stuff will do absolutely nothing," he said. "There's no signal there. However, if you want to say turn right, then your brain will send the signal to your vocal chords, which we pick up."

So it takes a little practice learning how to get the job done, but Coleman has learned how to control the wheelchair without moving his hands by simply thinking about saying the words that the sensor can detect and send to the computer, where they are translated into commands.

But that's not the immediate goal of the pair.

"The goal of our technology is not to move a wheelchair," Callahan said. "The goal of our technology is to allow people to speak."

According to the U.S. Census Bureau, there are about 2.5 million people in the United States who have limited speech, and about 500,000 who cannot speak at all. Globally, Callahan said, "about 60 million people worldwide have either limited or no form of speech."

So there are a lot of folks out there who could benefit from this technology. Let's hope it works as well in the real world as it does in the lab.

Lee Dye is a former science writer for the Los Angeles Times. He now lives in Juneau, Alaska.