Paralyzed people driving electric wheelchairs using only their minds

A new brain-machine interface (BMI)-controlled electric wheelchair allowed quadriplegic people to successfully navigate an obstacle-filled room. The results of this trial were published in iScience, and they represent a fairly successful trial of this kind in quadriplegic subjects.

The University of Texas at Austin researchers reported that, by the end of the study, two of the three subjects moved their BMI-controlled wheelchairs with 95% to 98% accuracy. They also reported that, at the beginning of the trial, these same subjects started at an accuracy of between 43% and 55%. 

How does a brain-machine interface work?

A BMI uses AI technology to act as an intermediary between the brain and the responsive machine. 

When people suffer injuries to their central nervous system that leave them living with quadriplegia, the brain loses the ability to tell the body what to do—especially how to move. However, by harnessing the still-active power of the brain, BMIs can facilitate movement for the subject instead.

In this study, an electrode cap worn on a subject’s head amplifies the brain’s electrical signals, which AI software then translates into commands for the wheelchair. 

"It works a lot like riding a horse," study co-author José del R. Millán said in a press release. "The rider can tell the horse to turn left or to go into a gate. But the horse will ultimately have to figure out the optimal way to carry out those commands."

The researchers attribute the trial’s success to both a) improvements in the AI and b) the subjects’ thoughts, which the researchers noted seemed to change as they spent more time with the device.

What else makes this particular study significant?

This trial was not the first of its kind. Previous relevant advancements included the BMI operation of a neurally-controlled robotic arm, an exoskeleton neuroprosthesis, and even other wheelchairs. 

This study’s use of a BMI also represents the effective use of a noninvasive method, the electrode cap, as opposed to a riskier, invasive interface that connects directly to the brain. These more invasive techniques are approaching safety and feasibility for human trials, but they aren’t there yet. 

Abbey Sawyer, a researcher familiar with the study’s work, told US News: “This is probably the first small study to achieve quite good success without having to enter the brain. There’s much more invasive approaches which are entering safety and feasibility stages of human trials at this point, but this is one of the first and probably one of the most successful of a noninvasive approach."

This study was also one of the first to only conduct tests with tetraplegic subjects. No able-bodied subjects were enrolled in the trial. 

What’s next for brain-controlled mobility devices?

We certainly hope that this life-changing tool becomes available for disabled people in the near future. The researchers are optimistic that this study points toward the future commercial viability of noninvasive mind-controlled wheelchairs.

Millán said: “We demonstrated that the people who will actually be the end users of these types of devices are able to navigate in a natural environment like their hospital with the assistance of a brain-machine interface.”

Of course, part of developing a feasible solution is finding workarounds for the limitations at play. In the case of this mind-controlled wheelchair, the trial revealed that several options will be needed for the device to help more people. While two out of three of the subjects significantly improved their accuracy in operating the device, the third did not show these improvements.

This result also underscores the need to prioritize testing these solutions with the people who will use them, rather than with able-bodied people. We look forward to seeing more devices of this kind developed and tested with the people who will benefit from them.