Brain-computer interface technology is advancing rapidly, but it currently relies on wires that seriously limits its use in everyday applications. That could soon change, though, as researchers recently completed the first human trial of a high-bandwidth wireless neural interface.
The most accurate way to record brain signals today is by using a device called an intracortical brain-computer interface (BCI), which involves an array of electrodes being implanted into a patient’s motor cortex. Signals from these electrodes then pass to a port in their skull, which connects to cables that transmit the signal to an external computer.
The highly invasive nature of the implantation procedure means the devices are still only used for research in a very small number of patients. But there’s been major progress in the kinds of things users have been able to accomplish using these devices, from typing on computers to controlling robotic prosthetics and even moving paralyzed limbs.
But the fact that users need to be physically wired into these systems seriously limits the activities they can perform, as well as researchers’ ability to test them over long periods of time and in diverse settings. Now though, a team from Brown University has shown that a wireless BCI can record brain signals with the same fidelity as a wired device for up to 24 hours in a patient’s home.
“We’ve demonstrated that this wireless system is functionally equivalent to the wired systems that have been the gold standard in BCI performance for years,” study leader John Simeral, from Brown University, said in a press release.
“The only difference is that people no longer need to be physically tethered to our equipment, which opens up new possibilities in terms of how the system can be used.”
The study, reported in IEEE Transactions on Biomedical Engineering, builds on a prototype wireless transmitter designed by Brown engineers in 2014. The system was designed to work with a wired brain-computer interface called BrainGate, also developed at Brown, which relies on two 96-electrode arrays implanted beneath the patient’s skull.
The revamped transmitter is about two inches across, and can be connected to the same port used by the wired system’s cables. The unit digitizes the recorded brain signals and then transmits them to a series of antennae positioned around the user’s room.
To demonstrate the potential of the system, the researchers showed that two patients who had been paralyzed by spinal cord injuries were able to use the device to control a computer cursor in their homes rather than in a specialized lab. They also showed that it was possible to record one of the patient’s neural activity for 24 hours straight thanks to the device’s 36-hour battery life.
This isn’t the first demonstration of a wireless BCI, but previous devices have been lower bandwidth than the gold standard wired systems. The new device matched the fidelity of the wired system while removing the need for patients to be tethered to a computer, which the researchers say could open up a host of new possibilities.
“The evolution of intracortical BCIs from requiring a wire cable to instead using a miniature wireless transmitter is a major step toward functional use of fully implanted, high-performance neural interfaces,” said study co-author Sharlene Flesher, a hardware engineer at Apple who was at Stanford University when the research was conducted.
As well as enabling researchers to tackle a broad swathe of new questions in neuroscience, the authors hope the device could eventually help restore the independence of people suffering from paralysis. The breakthrough is also likely to pique the interest of companies like Neuralink and Kernel who hope to one day make neural interfaces standard consumer technology.
The bulky nature of the transmitter, complicated receiver setup, and invasive procedure required to install the implants are major hurdles, but the research is a significant step to making BCIs a viable technology for everyday activities.
Image Credit: Raman Oza from Pixabay
* This article was originally published at Singularity Hub
0 Comments