Battelle’s NeuroLife uses thought to control paralyzed limbs

Patient Ian Burkhart (front) and project electrical engineering lead Nick Annetta (back right). (photo by Battelle)

Researchers at Battelle today published a paper in Nature covering the progress of its NeuroLife system, implanted in quadriplegic patient Ian Burkhart. In 2014, Burkhart became the 1st patient ever to achieve motor control of his paralyzed arms and hands through his own thoughts.

The NeuroLife system restored enough motion for Burkhart to perform tasks such as pouring out bottles, mixing liquids and taking a credit card from his wallet and swiping it. Burkhart, a quadriplegic, was injured in a 2010 diving accident that paralyzed him from the chest down.

The NeuroLife system, which is the culmination of many years of work for the researchers and engineers at Battelle, consists of a brain-implanted microchip, a computer system with custom built algorithms and machine learning, and specially made sleeves designed to stimulate arm muscles – allowing the patient use thoughts to control the limb.

“The project really got under way about 6 years ago, in the development of the system we have today,” electrical engineering task lead Nick Annetta told MassDevice.com yesterday. “A lot of that development and planning was really working to try and be ready for the different type of signals we thought we might see in the brain.”

To develop the system, Annetta and his Battelle team collected as much as a gigabyte of data every minute from the chip implanted in Burkhart’s brain. The diminutive 4mm by 4mm chip, a Utah array with 96 electrodes, is designed to pick up small changes in voltage caused by neurons firing near the electrodes, he explained.

“So, that data goes through a computer that runs our algorithms. It takes that data, compresses it, runs some processing on it and runs it into our machine-learning algorithm. That decodes what [Burkhart]’s thinking about, what motion he’s thinking about. And then we have another method that we use to figure out what pattern is needed to stimulate the forearm to invoke the movement,” Annetta told us.

The computer then decides what pattern to stimulate and transmits control commands to the neuromuscular electrical stimulator attached to Burkhart’s arms. The stimulation sleeve – a feat in and of itself, Annetta said – was designed to generate pulses that are then fed through the forearm cuff to invoke motion.

“There’s nothing on the market like it, so we had to design our own,” he said.

Similar stimulation devices may have as many as 8 channels, Annetta added; the sleeve Battelle developed has 160. And the sleeves hold the potential to do more than help restore motion for paralyzed individuals.

“That sleeve could be used in a lot of applications, not just this particular system. It could be used for stroke rehabilitation without having to have an implant in the brain. That’s just 1 example,” Annetta said.

Researchers on the team spent about a month poring over data from the chip before Burkhart was connected to the stimulation sleeves, and spent a significant amount of time prior to that learning to move a virtual hand on a screen, Annetta told us.

The virtual hand served 2 purposes, Annetta explained. The 1st was to cue Burkhart on what motions he should be thinking about – for example, displaying a particular hand motion, such as opening or closing, or pointing an index finger. Burkhart would then be instructed to concentrate on making that same motion with his hand, entirely in his mind.

After they had used the cue hand to collect information from the sensor on what neurons had fired while he was concentrating on the hand, Burkhart was provided with a 2nd, feedback hand, “which basically presents to him what we think he’s thinking about – what we’ve decoded that he’s thinking about,” Annetta explained.

The virtual hand allowed the team to see if the system would work without having to electrically stimulate the forearm – something Annetta said would introduce another major hurdle.

“This was kind of a simplified version of the system. It not only allowed us to check and see if it would work, but it also allowed us to see how many motions we could really control all at once. That’s 1 thing that kind of separates this kind of system from something that might be controlled from EEG,” he said.

The NeuroLife system allows for very precise control, such as flexing the fingers separately or together, a feat Annetta said is much more delicate than an EEG would be able to accomplish.

After mastering the virtual hand, Burkhart was moved to the real thing. Annetta credited Burkhart for remaining optimistic in the face of years of hang-ups and delays as they worked the kinks out of the system.

“He never really showed signs of despair or any real negativity at all, he just seemed to want to keep going. He was always hoping it was going to improve and having faith that everyone was working toward the same goal, and [that] we’d make it work, we’d advance the field so someone could make use of this in the future,” Annetta said.

After months of work once the prototype was ready, Burkhart was able to operate an arm and hand using the system. Eventually, he was able to master many motions, such as gripping a bottle, pouring out the contents, setting it down and grasping a tool to stir what he’d just poured. Burkhart was able to play basic video games, hoist a variety of mugs, lift a phone to his ear and – importantly for Burkhart – take a credit card out of his wallet and swipe it.

“This was an idea he presented to us. We asked him what would be relevant to him in his everyday life. He said that was a motion he had lost that would be significant to him to regain,” Annetta told us.

The restoration of motion to Burkhart’s arms is somewhat bittersweet, he added, as the device is not portable enough to leave the lab.

“This is the type of work that I’ve always wanted to do, and for a very long time I’ve wanted to work on something like this. To see it actually happening is incredible, but at the same time, Ian can’t take the system home with him. So, he leaves it in the clinic [every day],” Annetta said. “Working with Ian is an absolute privilege. There’s no way I personally expected to have a participant anything like Ian. The amount he gives to the program is incredible.”

Battelle is seeking more participants for the NeuroLife program, he said. Much more work is required to develop a commercialized device, Annetta added.

“It’s invigorating, but at the same time, it makes you really want to buckle down and work harder.”

The post Battelle’s NeuroLife uses thought to control paralyzed limbs appeared first on MassDevice.

from MassDevice http://ift.tt/1WrYvIQ