February 15, 2017

McGuire VAMC Develops BCI-controlled Prosthesis

Content provided by The O&P EDGE
Current Issue - Free Subscription - Free eNewsletter - Advertise

William Gadsby, a U.S. Marine Corps veteran from California, has been participating in the testing of the BCI. Photography courtesy of Hunter Holmes McGuire VAMC.

Douglas Murphy, MD, with Hunter Holmes McGuire VA Medical Center (VAMC), Richmond, Virginia, is working with ground-breaking technology to allow veterans to control prosthetic legs using only their thoughts. Using a brain-computer interface (BCI) patients are able to activate a switch that unlocks a prosthetic knee, instead of making manual adjustments. This system offers improved mobility and affects activities like standing up, sitting down, and walking. Murphy presented the technology, which is in the early stages of development, at the Association of Academic Physiatrists (AAP) Annual Meeting in Las Vegas on February 10.

The BCI allows the prosthetic user a quick, hands-free system to control his or her prosthesis in environments with varying terrains, and to expend less energy than is typically required. The system is easy to use and learn. The person is trained to use a specific thought, such as flexing a knee, for control. The thought generates electrical activity in the nerve cells and brainwaves. A chip can be implanted in the brain to pick up electrical activity or electrodes can be placed on the scalp to pick up brainwaves. In people with paralysis or amputations, BCIs can be used to control the movement of muscles, limbs, and prostheses.

“In general, using a prosthesis is an unnatural act that requires training, extra effort, and can have a certain amount of awkwardness to it,” explained Murphy, who is also an associate professor of physical medicine and rehabilitation at Virginia Commonwealth University and lead investigator (along with Ou Bai, PhD, from Florida International University) of a recent study that sought to establish the feasibility of manipulating a prosthetic knee with a BCI.

In the study, Murphy’s team worked with a person with a transfemoral amputation. Using surface scalp electrodes to transmit brainwave data to a computer software program, the participant was taught how to activate a knee-unlocking switch through mental imaging.

“In our first attempt at using [a] BCI with a lower-extremity prosthesis, we wanted to test a simple system before moving on to more complicated ones to test the feasibility of the concept,” said Murphy. “Thus, we chose control of the simplest prosthetic knee, which is the manual locking knee. When locked, the knee is rigid and straight, and when unlocked the knee swings freely. Someone with an above-knee amputation would have to physically/manually unlock the knee to sit and could lock or unlock in standing or walking, depending on his or her needs. We were interested to see if our participant could literally ‘think’ his way to unlocking his prosthetic.”

The participant learned to activate the knee-unlocking switch on his prosthesis that turned on a motor and unlocked his prosthetic knee. He then proceeded to walk back and forth in parallel bars while demonstrating his ability to unlock the knee to swing his leg and to sit down. Throughout the study, the participant was able to successfully unlock his knee 50 to 100 percent of the time.

“The ultimate goal of this research is to provide the individual with a prosthesis that more easily and more successfully meets his or her needs for movement and walking,” Murphy said. “The system should be comfortable, easy to use, and serve useful purposes. The patient’s subjective experience should reflect these goals. Our subject gave a very good example of how this system could help him. He likes to hike with his children…. With BCI control he could adjust his prosthesis for descending the hill very easily. This is the type of daily life activity we believe can be improved with [a] BCI.”

Editor’s note: This story was adapted from materials provided by Hunter Holmes McGuire VAMC and AAP.

Bookmark and Share