A new brain-computer interface (BCI) allowed three individuals with severe physical disabilities to type words using only their thoughts at speeds four times faster than what was previously possible, according to a new study published in eLife.
Researchers have been pursuing BCIs for years because of their potential to restore function to people with disabilities, but
, "few devices are ready for widespread practical use," Scientific American reports. BCIs used for typing, for example, have been held back by slow typing speeds.
But a new BCI created by researchers at Stanford University and other members of the school's BrainGate consortium, which includes Massachusetts General Hospital, enables paralyzed and other disabled individuals to type faster than ever before.
The study focused on two individuals with amyotrophic lateral sclerosis (ALS) and another who was paralyzed after an accident. Each participant had one or two surgically implanted "intracortical" sensors that contained microelectrodes that went 1 to 1.5 millimeters into their brain's motor cortex. Those electrodes transmitted neural signals through a cable to a computer that used special algorithms—developed by the researchers—to translate brain signals into commands that moved a cursor on a screen. With "minimal training," according to a Stanford release, each immobile patient learned to type using the cursor.
To see how fast each participant could type using the system, the researchers had them complete three tasks:
- Free typing, which involves answering questions;
- Copy typing, which involve typing out set phrases; and
- Selecting squares on a grid as they lit up.
Each task was a useful way of measuring the BCI's utility, the Scientific American reports, but the copy typing and the square-selection task are considered more standardized measures of typing speed. The square-selection task is particularly useful because it can be measured in digital "bits per second," or the maximum speed at which the system can output information, according to Scientific American.
The two participants with ALS typed at 2.2 and 1.4 bits per a second, which was more than double the previous record. The paralyzed individual achieved 3.7 bits per second—four times faster than what was previously possible.
On the free-typing task, the two individuals with ALS achieved between about three- and six-words per minute. The individual with paralysis achieved almost eight, according to Scientific American. Each participant posted similar speeds on the copy-typing test. The researchers also say that these speeds could increase if the software used predictive algorithms to complete words before they are fully types.
Co-lead author Chethan Pandarinath, then a postdoctoral fellow at Stanford, stressed that the study was important both because it introduced new technology and also helped establish standards to measure progress in the future. "We need to establish measures so that—in spite of potential variability between people, methods, and researchers—we can really say, 'Clearly this new advance led to higher performance,'" he said, citing the team's decision not to use predictive software.
Other researchers not involved with the study said it was significant because it showed progress toward making BCIs valuable tools for patients. Biomedical engineer Jennifer Collinger of the University of Pittsburgh said, "This is the fastest typing anyone has shown with a BCI. It's on par with technologies like eye-trackers, but there are groups those technologies don't work for."
Andrew Schwartz also of the University of Pittsburgh added that the team's BCI was superior to other less-invasive BCI systems such as lectroencephalography. Even though movement and scarring reduce the effectiveness of the brain implants over time they are still "much better than you get with any other technique," he said.
Next, the researchers want to work toward making the sensors wireless, which Pandarinath says could take five to 10 years. "That's a critical step [toward] a device that you could send somebody home with and be less worried about potential risks like infection," he explained.
The researchers also said technology and algorithms that allowed the study participants to use a cursor to type—essentially a "point and click" action—could be applied to a variety of computing devices, such as smartphones or tablets, with few modifications. Pandarinath said this point-and-click interface could "open a whole new realm of function beyond communication: surfing the Web, playing music, all sorts of things able-bodied people take for granted."
Neurosurgeon and co-senior author, Jaimie Henderson of Stanford, said, "One of our main goals is to allow 24 hours a day, seven days a week, 365 days a year control of a standard computer interface using only brain signals" (Makin, Scientific American, 2/21; Stanford University release, 2/21; Thompson, Popular Mechanics, 2/21).
The neuroscience clinical technology compendium
There's no shortage of neuroscience clinical technologies available for your service lines, so how do you choose the right one for your organization? To maximize the cost-effectiveness of your investments, you'll need to make smart choices.
In this compendium, we provide overviews of emerging clinical technologies—for neurology, neurosurgery, neuroimaging and more—as well as considerations for investment and keys for investment success. Download it now and learn the right technologies for gaining new capabilities, improving clinical effectiveness, and differentiating your organization from the competition.