The December 2022 issue of IEEE Spectrum is here!

Close bar

Quadriplegic Pilots Race for Gold in Cybathlon Brain Race

Close finish in the brain-computer interface race at world's second cyborg Olympics

3 min read
A computer generated car moves down an elevated road. An inset shows a masked man wearing a cap studded with electrodes.
Francesco Bettella, pilot for the Italian team, finished in 2 minutes 52 seconds for the win.
Image: Cybathlon/ETH Zurich

The competitors were neck-and-neck going into the final turns of the last heat, and in the end, Italy beat Thailand by four seconds. But unlike the Olympic games, none of the competitors in this race could move their bodies. Instead, they competed using only their thoughts. 

This is Olympic racing, cyborg-style. Using brain-computer interface (BCI) systems, the competitors—all of whom are paralyzed from the neck down—navigated computer avatars through a racetrack using thought-controlled commands. 

The race was part of Cybathlon 2020: The second ever cyborg Olympics, in which people with paralysis or amputated limbs turn themselves into cyborg athletes using robotics and algorithms. Proud competitors raced with their exoskeletons, powered wheelchairs and prosthetic limbs through obstacle courses as their tech teams cheered them on.

Unlike the last Cybathlon, in 2016, where teams competed live in front a sold-out arena, this year’s organizers weaved together remote events in a pandemic-conscious way. Teams from around the world built obstacle courses in their labs and recorded their races in front of an official referee. Cybathlon organizers then integrated the recordings into a live webcast, complete with commercials and a veteran Olympic commentator.

For the BCI race, each pilot donned an electrode cap that read the electrical activity of his brain using electroencephalography (EEG). Algorithms interpreted their brain signals as commands, allowing the pilots to direct an avatar through a computer game. 

The game consisted of a race track with straights and turns that pilots had to navigate, and sections in which they had to turn on their headlights. Pilots controlled the avatars by thinking about—but not actually making—physical movements. For example, the avatar would turn left if the pilot imagined moving his left arm, and right if the pilot imagined moving his right arm. Thinking about a movement such as lifting both feet could turn on the headlights. 

Some of the hardest parts of the track were the straightaways, says Nicole Wenderoth, professor of neural control of movement at ETH Zurich, whose team entered its BCI system in the event. For those sections, pilots had to concentrate on thinking about nothing. 

“You can try to think ‘left’ harder. But when you're supposed to do nothing, and the algorithm isn’t getting it, you have to try to do nothing harder,” she says. That’s extremely difficult to do, particularly in a competition setting where emotions are involved. And it’s the opposite of just about any other competition, where one typically tries hard to do something. 

Piloting Wenderoth’s team’s BCI system was Samuel Kunz, a design engineer who describes himself as “quite competitive.” He suffered an accident six years ago that left him paralyzed in all four limbs. Kunz finished fourth in the race with a best time of 3 minutes, 41 seconds.

The winner of the event, an Italian team piloted by Paralympic swimmer Francesco Bettella, finished in 2 minutes 52 seconds. “The secret is to stay calm and focused and not be surprised by the agitation happening around you,” because it can translate into false commands to the avatar, Bettella said in a recorded interview during the webcast

How much BCI technology has improved since the 2016 Cybathlon games is hard to quantify, since the computer games used in the two competitions were different, says Wenderoth. “Subjectively, I had the impression that certainly the top pilots in 2020 did better than the top pilots in 2016,” she says.

Perhaps that’s because the EEG has become more stable in noisy environments, or because there’s more knowhow in decoding brain activity in people with quadriplegia. She adds: “In theory, there has been a big leap on the algorithmic side, which is all related to deep learning and different types of networks that you can use. However, my gut feeling is telling me that probably the simpler algorithms are the more successful ones” for this event, she says. 

And that’s both the challenge and the purpose of the Cybathlon: to encourage engineers to create BCI and other cyborg systems that are simple enough and reliable enough for people to use on their own in a real-world setting; to get these technologies out of the lab and into the world where they can actually help people. 

For now, BCI control seems to be particularly challenging for pilots with severe impairments, says Wenderoth. The BCI race winner, Bettella, is an excellent BCI pilot and also has some ability to move his arms as a Paralympic swimmer. Bettella of course did not use his arms during the race, but the surviving neural connections between his brain and his arms may have helped him. 

In fact, it would be considered cheating if a team were to incorporate any movement, particularly eye or face movements, into their algorithms. (In true Olympic fashion, an official Cybathlon referee was on site with each cyborg competitor to help ensure no cheating occurred.) “If you activate a muscle in your face, you are creating signals in the range of millivolts,” says Wenderoth. “That’s three magnitudes higher than what you're going to extract from the brain,” and a lot easier for EEG systems to detect and interpret, she says.

The Conversation (0)

Are You Ready for Workplace Brain Scanning?

Extracting and using brain data will make workers happier and more productive, backers say

11 min read
Vertical
A photo collage showing a man wearing a eeg headset while looking at a computer screen.
Nadia Radic
DarkGray

Get ready: Neurotechnology is coming to the workplace. Neural sensors are now reliable and affordable enough to support commercial pilot projects that extract productivity-enhancing data from workers’ brains. These projects aren’t confined to specialized workplaces; they’re also happening in offices, factories, farms, and airports. The companies and people behind these neurotech devices are certain that they will improve our lives. But there are serious questions about whether work should be organized around certain functions of the brain, rather than the person as a whole.

To be clear, the kind of neurotech that’s currently available is nowhere close to reading minds. Sensors detect electrical activity across different areas of the brain, and the patterns in that activity can be broadly correlated with different feelings or physiological responses, such as stress, focus, or a reaction to external stimuli. These data can be exploited to make workers more efficient—and, proponents of the technology say, to make them happier. Two of the most interesting innovators in this field are the Israel-based startup InnerEye, which aims to give workers superhuman abilities, and Emotiv, a Silicon Valley neurotech company that’s bringing a brain-tracking wearable to office workers, including those working remotely.

Keep Reading ↓Show less
{"imageShortcodeIds":[]}