Key points from article :
A groundbreaking brain-computer interface (BCI) has allowed a paralyzed man to control a robotic arm for a record seven months without major recalibration, marking a major step toward more reliable assistive technology. The research, led by neurologist Karunesh Ganguly at UC San Francisco, was published in Cell. Unlike previous BCIs, which typically function for only two days before requiring recalibration, this system adapts to natural shifts in brain activity over time, allowing for long-term use.
The technology works by implanting small sensors on the brain's surface to detect movement-related neural signals. Over time, the AI-driven model learns to adjust for daily fluctuations in brain activity, preventing the signal disruptions that plagued earlier BCIs. The study participant, who was paralyzed after a stroke, first trained on a virtual robotic arm before successfully using a real robotic arm to grasp and move objects. He later performed more complex tasks, such as retrieving a cup from a cabinet and placing it under a water dispenser.
Even months later, the patient retained his ability to control the robotic arm with just a brief 15-minute calibration session. Ganguly and his team are now working to refine the system for smoother, faster movements and hope to test it in real-world home settings. This breakthrough offers new hope for individuals with paralysis, bringing assistive robotics closer to practical, everyday use.
