In a stunning scientific breakthrough, researchers at UC San Francisco have enabled a paralyzed man to control a robotic arm—just by thinking about it. This cutting-edge achievement, made possible through a brain-computer interface (BCI), represents a major leap forward in neurotechnology and artificial intelligence.
The man, who had been paralyzed for years due to a stroke, was able to grasp, move, and drop objects solely through the power of his imagination. Unlike previous BCI systems that only worked for a short period, this revolutionary device functioned seamlessly for a record seven months, thanks to advanced AI adaptations.
The Power of the Mind
The BCI technology relies on an AI model that learns and adapts to subtle shifts in brain activity over time. As the participant repeatedly imagined making movements, the AI continuously adjusted, refining its ability to translate thought into action.
“This blending of learning between humans and AI is the next phase for these brain-computer interfaces,” said Dr. Karunesh Ganguly, the lead researcher and professor of neurology at UCSF Weill Institute for Neurosciences. “It’s what we need to achieve sophisticated, lifelike function.”
One of the key challenges in making BCIs work long-term has been the shifting nature of brain signals. Dr. Ganguly and his team discovered that while brain activity representing movement remains stable in shape, its location in the brain subtly shifts from day to day. This explained why previous BCIs would lose effectiveness so quickly. By programming the AI to account for these shifts, the researchers created a system that remained functional for months with minimal adjustments.
To train the AI, the study participant first imagined moving various body parts, such as his hands, feet, and head. Sensors implanted on the surface of his brain captured these signals, and the BCI recorded how his brain represented these movements.
Read Also: Robotic Companions to Offer Tailored Care for Elderly and Patients
From Virtual to Reality
Initially, the participant’s control over the robotic arm was imprecise. To refine his skills, researchers introduced a virtual robotic arm, allowing him to practice visualizing movements with real-time feedback. This helped train his brain to send clearer signals.
Once he transitioned to a physical robotic arm, the improvement was remarkable. With just a few practice sessions, he was able to pick up blocks, turn them, and move them to new locations. He even managed to open a cabinet, remove a cup, and position it under a water dispenser.
Months later, after a quick 15-minute recalibration session, he was still able to control the robotic arm effectively, proving the long-term viability of the system.
Dr. Ganguly and his team are now refining the AI to make the robotic arm’s movements faster and more natural. Their next goal is to test the BCI in a home setting, where it could provide real-world benefits to people with paralysis.
For millions living with mobility impairments, even simple tasks like eating or getting a drink of water remain a challenge. This breakthrough brings them one step closer to regaining independence.
“I’m very confident that we’ve learned how to build the system now, and that we can make this work,” said Dr. Ganguly.
With AI-driven BCIs evolving rapidly, a future where people with paralysis can control assistive devices with their thoughts is not just a dream—it’s becoming reality.
- Article Source: CELL
- Related Article: The Fulani’s True Origins, Solved by Science
- For Firsthand Information, Join the ScienceNaija channel on WhatsApp
