[ad_1]
× closure
Hussein Alawieh, a graduate student in the laboratory of Dr. José del R. Millán, wears a hat containing electrodes connected to a computer. Electrodes collect data by measuring electrical signals from the brain, and decoders interpret the information and translate it into game actions.Image source: University of Texas at Austin
Think of playing a racing game like Mario Kart and executing a series of complex turns in a lap using only your brain.
This is not a video game fantasy, but a real program created by engineers at the University of Texas at Austin as part of brain-computer interface research to help improve the lives of people with movement disabilities. What’s more, the researchers combined machine learning capabilities with their brain-computer interface, making it a one-size-fits-all solution.
Typically, these devices require extensive calibration for each user—every user’s brain is different, whether healthy or disabled—which has been a major barrier to mainstream adoption. This new solution can quickly learn the needs of a single subject and self-calibrate through repetition. This means multiple patients can use the device without the need for individual adjustments.
“When we think about this in a clinical setting, this technology will enable that so we don’t need a dedicated team to go through this long and tedious calibration process,” said Satyam Kumar, a graduate student in José’s lab. del R. Millán, the Chandra Family Professor in the Department of Electrical and Computer Engineering in the Cockrell College of Engineering and the Department of Neurology at Dell Medical School. “It’s faster to move from one patient to another.”
Research on calibration-free interfaces was published in Proceedings of the National Academy of Sciences.
The subjects wore hats containing electrodes connected to a computer. Electrodes collect data by measuring electrical signals from the brain, and decoders interpret the information and translate it into game actions.
Milan’s work on brain-computer interfaces can help users guide and enhance their neuroplasticity, the brain’s ability to change, grow and reorganize over time. The experiments aim to improve patients’ brain function and make their lives easier using devices controlled by brain-computer interfaces.
× closure
From left to right: Satyam Kumar, Hussain Alawiyeh and Jose Del R. Milan.Image source: University of Texas at Austin
In this case, the action is twofold: a racing game and the simpler task of balancing the left and right sides of the number bar. An expert was trained to develop “decoders” for simpler pub tasks, enabling the interface to convert brain waves into commands. The decoder serves as the basis for other users and is key to avoiding lengthy calibration processes.
The decoder worked so well that subjects were trained on both a pub game and a more complex racing game, which requires thinking several steps ahead to turn a corner.
The researchers called the work fundamental because it lays the foundation for further innovations in brain-computer interfaces. The project used 18 subjects without movement disorders. Eventually, as they continue down this path, they will test it on people with movement disorders and apply it to larger groups in clinical settings.
“On the one hand, we want to translate brain-computer interfaces into the clinical field to help people with disabilities; on the other hand, we need to improve our technology to make it easier to use and thus have a greater impact on these people with disabilities,” Millan said. .
In parallel with translational research, Milan and his team continue to work on wheelchairs that users can drive through a brain-computer interface. At this month’s South by Southwest conference and festival, researchers demonstrated another potential use for the technology, controlling two rehabilitation robots with hands and arms.
This is not part of a new paper, but rather a sign of where this technology is heading in the future. Several people volunteered and successfully operated the brain-controlled robot within minutes.
“The purpose of this technology is to help people, to help them with their daily lives,” Millan said. “We will continue on this path to help people wherever we go.”
More information:
Satyam Kumar et al., Transfer learning promotes the acquisition of personal BCI skills, Proceedings of the National Academy of Sciences (2024). DOI: 10.1093/pnasnexus/pgae076
Journal information:
Proceedings of the National Academy of Sciences
[ad_2]
Source link