ORIGINAL: IEEE Spectrum
By Eliza Strickland
11 Aug 2014
By Eliza Strickland
11 Aug 2014
A new brain-computer interface lets DIYers access their brain waves
Photo: Chip AudetteEngineer Chip Audette used the OpenBCI system to control a robot spider with his mind.
The guys who decided to make a mind-reading tool for the masses are not neuroscientists. In fact, they’re artists who met at Parsons the New School for Design, in New York City. In this day and age, you don’t have to be a neuroscientist to muck around with brain signals.
With Friday’s launch of an online store selling their brain-computer interface (BCI) gear, Joel Murphy and Conor Russomanno hope to unleash a wave of neurotech creativity. Their system enables DIYers to use brain waves to control anything they can hack—a video game, a robot, you name it. “It feels like there’s going to be a surge,” says Russomanno. “The floodgates are about to open.” And since their technology is open source, the creators hope hackers will also help improve the BCI itself.
Photo: OpenBCI The OpenBCI board takes in data from up to eight electrodes.
Their OpenBCI system makes sense of an electroencephalograph (EEG), signal, a general measure of electrical activity in the brain captured via electrodes on the scalp. The fundamental hardware component is a relatively new chip from Texas Instruments, which takes in analog data from up to eight electrodes and converts it to a digital signal. Russomanno and Murphy used the chip and an Arduino board to create OpenBCI, which essentially amplifies the brain signal and sends it via Bluetooth to a computer for processing. “The big issue is getting the data off the chip and making it accessible,” Murphy says. Once it’s accessible, Murphy expects makers to build things he hasn’t even imagined yet.
The project got its start in 2011, when Russomanno was a student in Murphy’s physical computing class at Parsons and told his professor he wanted to hack an EEG toy made by Mattel. The toy’s EEG-enabled headset supposedly registered the user’s concentrated attention (which in the game activated a fan that made a ball float upward). But the technology didn’t seem very reliable, and since it wasn’t open source, Russomanno couldn’t study the game’s method of collecting and analyzing the EEG data. He decided that an open-source alternative was necessary if he wanted to have any real fun.
Happily, Russomanno and his professor soon connected with engineer Chip Audette, of the New Hampshire R&D firm Creare, who already had a grant from the U.S. Defense Advanced Research Projects Agency (DARPA) to develop a low-cost, high-quality EEG system for “nontraditional users.” Once the team had cobbled together a prototype of their OpenBCI system, they decided to offer their gear to the world with a Kickstarter campaign, which ended in January and raised more than twice the goal of US $100,000.
Murphy and Russomanno soon found that production would be more difficult and take longer than expected (as is the case with so many Kickstarter projects), so they had to push back their shipping date by several months. Now, though, they’re in business—and Russomanno says that shipping a product is only the beginning. “We don’t just want to sell something; we want to teach people how to use it and also develop a community,” he says. OpenBCI wants to be an online portal where experimenters can swap tips and post research projects.
So once a person’s brain-wave data is streaming into a computer, what is to be done with it? OpenBCI will make some simple software available, but mostly Russomanno and Murphy plan to watch as inventors come up with new applications for BCIs.
Audette, the engineer from Creare, is already hacking robotic “battle spiders” that are typically steered by remote control. Audette used an OpenBCI prototype to identify three distinct brain-wave patterns that he can reproduce at will, and he sent those signals to a battle spider to command it to turn left or right or to walk straight ahead. “The first time you get something to move with your brain, the satisfaction is pretty amazing,” Audette says. “It’s like, ‘I am king of the world because I got this robot to move.’ ”
In Los Angeles, a group is using another prototype to give a paralyzed graffiti artist the ability to practice his craft again. The artist, Tempt One, was diagnosed with Lou Gehrig’s disease in 2003 and gradually progressed to the nightmarish “locked in” state. By 2010 he couldn’t move or speak and lay inert in a hospital bed—but with unimpaired consciousness, intellect, and creativity trapped inside his skull. Now his supporters are developing a system called the BrainWriter: They’re using OpenBCI to record the artist’s brain waves and are devising ways to use those brain waves to control the computer cursor so Tempt can sketch his designs on the screen.
Another early collaborator thinks that OpenBCI will be useful in mainstream medicine. David Putrino, director of telemedicine and virtual rehabilitation at the Burke Rehabilitation Center, in White Plains, N.Y., says he’s comparing the open-source system to the $60,000 clinic-grade EEG devices he typically works with. He calls the OpenBCI system robust and solid, saying, “There’s no reason why it shouldn’t be producing good signal.”
Putrino hopes to use OpenBCI to build a low-cost EEG system that patients can take home from the hospital, and he imagines a host of applications. Stroke patients, for example, could use it to determine when their brains are most receptive to physical therapy, and Parkinson’s patients could use it to find the optimal time to take their medications. “I’ve been playing around with these ideas for a decade,” Putrino says, “but they kept failing because the technology wasn’t quite there.” Now, he says, it’s time to start building.