Apr 1, 2025
Researchers at UC San Francisco have achieved a remarkable breakthrough in brain-computer interface (BCI) technology, enabling individuals with paralysis to control robotic devices through thought alone.
This innovation combines artificial intelligence (AI) with neuroscience, allowing a paralyzed man to manipulate a robotic arm by imagining movements, a feat that marks a significant milestone in restoring autonomy to people with severe motor impairments.
The device, known as a brain-computer interface (BCI), represents a fusion of advanced AI and neural engineering. BCIs have previously struggled to maintain functionality over extended periods, often losing effectiveness after just one or two days. However, the newly developed BCI has set a record by functioning seamlessly for seven months without requiring major adjustments.
The key lies in the AI model’s ability to adapt to subtle changes in brain activity over time. As individuals repeatedly imagine movements, the AI refines its understanding of these neural signals, enabling more precise control of robotic devices. Dr. Karunesh Ganguly, neurologist and professor at UCSF, emphasized that this adaptive learning between humans and AI is critical for achieving lifelike functionality in neuroprosthetics.
Dr. Ganguly’s research revealed that while the shape of brain activity representations remains consistent, their locations shift slightly from day to day. This discovery explains why previous BCIs quickly lost their ability to interpret neural signals accurately.
To address this challenge, Ganguly and his team studied a participant who had been paralyzed by a stroke years earlier. Sensors implanted on the surface of his brain recorded neural signals as he imagined movements like grasping or lifting objects. Over two weeks, these signals were used to train the AI model to account for daily shifts in brain activity patterns.
Initially, the participant practiced controlling a virtual robotic arm that provided feedback on his imagined movements. This training helped refine his ability to visualize precise actions. Once he transitioned to using a real robotic arm, he quickly mastered tasks such as picking up blocks, opening cabinets and even holding a cup under a water dispenser.
Months later, the participant retained his ability to control the robotic arm with minimal recalibration, highlighting the long-term reliability of this BCI system. READ MORE
Stay Awake. Keep Watch
SOURCE: Fox News