A computer system that translates electrophysiologic signals and sends messages to an electrical stimulator is enabling a patient with a spinal cord injury (SCI) to walk over-ground, in a new case report.
The brain-computer interface (BCI) bypasses the damaged spinal cord to send signals directly to leg muscles.
The work is a culmination of 5 or 6 years of research, said study author An H. Do, MD, assistant professor, neurology, University of California at Irvine.
In an earlier paper (J Neuroeng Rehab. 2013;10:111), Dr Do and his colleagues showed that an able-bodied person and one with an SCI could use their electroencephalogram (EEG) signals to control a robotic gait exoskeleton suspended over a treadmill. When these participants thought about walking, a computer algorithm detected the EEG changes, triggering the robotic exoskeleton to generate walking movements. And when they stopped thinking about walking, the computer program detected the changed EEG signals and shut off the robotic exoskeleton.
Subsequent to these positive results, the researchers felt it was reasonable to progress to testing the concept of using a BCI system for over-ground walking.
"We saw incremental successes along the way and we were getting more confident," Dr Do told Medscape Medical News. "It was exciting to see it come together the way anticipated it to, or hoped it was going to pan out."
In the new paper, also published in the Journal of NeuroEngineering and Rehabilitation, the researchers describe the case of the same spinal cord injury patient as in the first paper — a physically active 26-year-old man who sustained a T6 AIS B injury.
This patient was chosen for the study because he passed all the stringent inclusion/exclusion criteria — having a T6-T12 injury and no medical complications, such as autonomic dysreflexia, orthostatic hypotension, pressure sores, peripheral neuropathy, severe osteoporosis, or lower-extremity fractures.
No Motor Function
The patient had no motor function in the lower extremities and no sensation below the injury level except for minimally preserved bladder fullness sensation.
He was trained for several weeks to learn to use his brain waves to control movements in a virtual reality environment.
The researchers used a character in a commercially available video game. "The brainwaves change as they're supposed to change when you're trying to move and those are the signals that are used to control the video game character," explained Dr Do.
The patient's muscles, which had atrophied because of several years of disuse after his paralysis, were then reconditioned.
To use the system, the patient wears a portable EEG amplifier in a backpack that sends wireless signals to a desktop computer. When the patient is standing up and starts thinking about walking, the system goes into gear.
The computer "does the number crunching" and sends wireless commands to the functional electrical stimulation system (FES). The EEG signals are recorded in real time and the computer analyzes the EEG in real time.
"As soon as it detects from the EEG that the patient is thinking about walking because the patterns change in a predictable manner, the computer sends a signal over to the electrical stimulation device and says, ok, turn on," said Dr Do.
The system is preprogrammed to stimulate the right leg to take a step and then the left leg. "It keeps alternating between the right and left leg until he stops thinking about walking, and at that point, the EEG goes back to the default idling state. Then the computer sends a signal to the electrical stimulator to stop that walking stimulation and keep his legs in a standing position."
For the current experiment, the patient had physical support to remain standing. His lower back muscles are weak from the injury and he can't keep his body upright as well as an able-bodied person, explained Dr Do.
The experiment demonstrated that the patient could use his brain power to stand still, initiate walking, shut off the walking pattern, and stand still again for several seconds. In total, he walked 3.8 m, in two 1.8-m sessions.
Although his brain waves are being used to ultimately control his leg movements, he is able to listen to and hear other people while he's taking steps, said Dr Do. "We never tested this formally, but it seemed that every now and then, he could say a couple words to us without interrupting the function of the system."
This might be because the computer program was developed to detect signals mostly in the sensory motor areas of the brain, he added. "Those other auditory or visual signals don't really register as the expected changes, so they are kind of ignored by the program."
According to Dr Do, studies in this field show that about 80% of people can be trained to use a brain-computer interface system.
One of the next steps for the researchers is to try to streamline the hardware, so it's less bulky and to start to work on specialized electronics that will go into a fully implantable BCI system, said Dr Do.
"The electrodes would probably come out of the skull and the cable would tunnel down to the chest wall. Our anticipation is that the computer unit would be implanted much like a deep-brain stimulator or a pacemaker in the chest wall."
In addition to patients with paraplegia, the system might be useful for those with tetraplegia and for those with stroke, said Dr Do.
For amputees, there is already technology being perfected using electromyography to control prosthetic limbs. "So for that group of people there is a pretty robust methodology already."
New Application of BCI
Asked to comment, Professor William McKinley, MD, director, Spinal Cord Injury Medicine, Virginia Commonwealth University Medical Center, Richmond, said that while BCI has been applied to other functional areas for patients with SCI, associating it with FES and walking capability is new.
"This area is quite interesting. It certainly represents an area of great interest for those with SCI," he told Medscape Medical News.
But although the study is "innovative" and "conceptually very intriguing" and demonstrates "feasibility," the technology is still at a very early stage in terms of its potential functional use. "It should not be considered a breakthrough yet," said Dr McKinley.
He stressed that this is a case report on only one individual. And he noted potential issues that may limit the technology's practicality. For example, over-ground walking — whether BCI controlled or not — requires a high level of energy expenditure.
"It's very tiring and many individuals choose to abandon these approaches in lieu of using a wheelchair."
As well, said Dr McKinley, some patients might view BCI-FES as a potentially cumbersome procedure, involving implantation, lengthy training, and the need for precision calibration.
"Nonetheless, this type of groundbreaking research should be embraced. We await further work on this front, combined with other advancements in BCI and FES technology as a means of achieving enhanced functional capabilities for those with SCI."
The authors and Dr McKinley have disclosed no relevant financial relationships.
J Neuroeng Rehabil. 2015;12:80. Abstract
Medscape Medical News © 2015 WebMD, LLC
Send comments and news tips to email@example.com.
Cite this: Paralyzed Patient Uses Brain-Computer Interface to Walk - Medscape - Oct 02, 2015.