Brain power

John P. Donoghue did not set out to cure anyone. He just wanted to explore the neural nitty-gritty of how the brain controls the body's actions, in everything from the lowly opossum on up to humans.

But his work may soon help the severely paralyzed.

Donoghue, 54, is the chief scientific officer of CyberKinetics Inc., a Foxborough-based company that is hoping to win approval from the US Food and Drug Administration any day now to begin testing a "brain-computer interface" device in a handful of paralyzed patients.

If the company's "BrainGate" device works as well in people as it has in monkeys, it will allow the patients to control a computer cursor simply with their thoughts, channeled by an array of 100 tiny electrodes that pick up signals directly from the brain. Such cursor control eventually could translate into such tasks as browsing the Web, typing, and controlling a robot limb.

Though not the first to explore the problem, Donogue is the scientific leader of the first company to try to commercialize a brain interface and make it broadly available for paralyzed patients. CyberKinetics plans to implant the device in up to five patients in the first trial and, if all goes well, to bring it to market in several years.

In 2002, Donoghue and his lab published a widely acclaimed paper in the journal Nature showing that by recording just a few neurons in a monkey, the animal could immediately control a cursor.

Now, if BrainGate works, Donoghue's team "will be, as far as I know, the first ones to put in a substantial number of electrodes and have a chance of reproducing in people what's been done in monkeys," said Dr. Jonathan R. Wolpaw, chief of the Laboratory of Nervous System Disorders at the Wadsworth Center, part of the New York State Department of Health, and a pioneer in brain-computer-interfaces.

Donoghue, who toiled for so long on the most basic science with little prospect of directly affecting people's lives, welcomes the twist in his career: "I can't imagine anything more fulfilling than to say, `I tried to understand how the brain works and I ended up helping paralyzed people.' "

But he has to acknowledge, he said, that as he was growing up in the Arlington home of his father, a bricklayer, and his mother, a homemaker, "I wasn't one of these people who, at 13, says, `I'm going to figure out how to make paralyzed people walk.' " The brain's mysteries captured Donoghue when, fresh out of his undergraduate years at Boston University, he was working with researchers who happened to be based on the grounds of the Fernald School in Waltham, home to many severely handicapped children. He would spend his days examining brains, he said, and walk outside to see children whose suffering was the result of brain disorders.

For most of his career, Donogue has worked to understand the motor cortex, the part of the brain's rind-like surface that controls motion. And the principal tool for such work has long been a tiny electrode implanted next to a brain cell to record its activity.

The ability to monitor a single neuron was a great advance more than a generation ago, but it also has a great flaw: Most agree that to understand the brain's function, they need to see the workings of groups of cells -- rather than isolated ones.

"Studying cells one at a time is like listening to a symphony when you only get to hear one instrument at a time," Donoghue said. "It's hard to understand the whole symphony."

Donoghue's journey from basic laboratory scientist to medical entrepreneur was spurred by his frustration with that limitation.

For years, he struggled to get around the "symphony" problem, but finally, in the mid-1990s, he decided in exasperation to scour the ranks of researchers at the annual meeting of The Society for Neuroscience, a city-sized gathering of nearly 30,000 researchers, for an engineer who could help him monitor the activities of more than one cell at a time.

He found Richard Normann, a University of Utah bio-engineer who had devised an array of electrodes on a platform about the size of a baby's fingernail. Normann aimed to use this "Utah Electrode Array" to help blind people regain some rudimentary vision. He thought a camera could translate light into electrical signals that could then be transmitted into their brains through the array, allowing them to "see" a bit.

The array was downright elegant, Donoghue said, but at that point had only been tried in lab dishes and cats.

So he and his colleagues undertook the trick of adapting it for use in primates. Quickly enough, they found that not only could the array read out the monkeys' brain signals, but that with recent advances in computing power, it could do it more or less in real time.

It worked "incredibly better than expected," Donoghue said.

Nonetheless, the team spent five more years working out the kinks.

Finally there came a moment in a hallway: Donoghue and his postdoctoral fellow, Nicholas Hatsopoulos, paused to talk and, as both recall it, discussed the notion that this elegant little device might actually be able to help humans, and that perhaps they should start a company to try.

The idea was a departure for Donoghue; he is the consummate academic -- chairman of Brown University's Neuroscience Department; holder of an endowed professorship; author of dozens of journal papers. But Hatsopoulos said he had long thought Donoghue would make a good chief executive officer: "A lot of other academics are prima donnas and would have a very hard time in business," he said, but Donoghue's easy people skills and talent for avoiding micromanagement made him a natural.

Donoghue said it was a mark of his business naivete that he thought of founding a company as a radical idea. It took 30 people working a full year to complete the necessary applications to the FDA for human clinical trials -- and it now seems utterly obvious that only a company could perform such a feat, he said.

Donoghue doesn't like "to use the `first' word," he said, and readily acknowledges that his lab is just one of a handful that has scored success in brain-computer interfaces in monkeys. And BrainGate won't be the first try at a direct brain-machine interface in humans. Such attempts go all the way back to some maverick scientists in the 1960s, he said, and an Atlanta doctor-scientist, Philip Kennedy, successfully implanted a wireless computer connection in a stroke patient's brain in 1998, allowing the patient to control a cursor.

The brain-interface field is "wide open now," Wolpaw said, and three different approaches all look promising. One picks up brain signals using electrodes on the scalp; the second implants electrodes inside the skull, on the surface of the brain; the third implants them in the brain tissue, as Donoghue does.

Wolpaw has focused his attention on the scalp version, and he and Donoghue have debated the relative merits of their contrasting positions. Wolpaw noted that implanted electrodes remain problematic in the long term, because the brain tissue reacts to the foreign object and scar tissue can form. Infection is also a threat.

But Wolpaw is looking forward to seeing what the BrainGate trials yield, he said.

"From knowing John and his work for a while, I think they will be done well."

When government-funded scientists first began developing brain-computer interfaces in the 1990, it wasn't at all obvious that the brain's signals could be turned into useful commands, said Dr. William Hettderks, who long oversaw the neural prosthesis program for the National Institute of Neurological Disease and Stroke. Now that has been shown, he said, and the next step is to determine whether "you can, in good conscience, develop a system that is safe enough, reliable enough, that you can justifiably put this into a person with a reasonable expectation that the benefit they might get out of this is worth the risk associated with the surgery." Donoghue's group, he said, is "ready to move on to the next step."

Donoghue describes with something like chagrin the many fawning calls he gets these days from television producers hoping for an exclusive on the first BrainGate patients. He tends toward the low-key and antihype.

But he also is not shy about proclaiming that BrainGate is just a small part of a dawning era of brain technology, one that will far outshine the scattered experiments of the past. "This is the age when we're really there," he said. "The technology is there. And it will be an enormous explosion -- I guarantee it -- of things with enormous health benefits."