Font size:
Print
Breakthrough in Brain-Computer Interface
Indian scientists help develop Brain-Computer Interface that enables movement for people who are paralysed
Context: Researchers at the University of California, San Francisco (UCSF) have developed a groundbreaking Brain-Computer Interface (BCI) that allows a paralysed man to control a robotic arm using only his thoughts.
More on News
- The lead author of the study, Dr. Nikhilesh Natraj, a neuroscientist and neural engineer who grew up in Chennai, India, is based at the Weill Institute for Neurosciences, UCSF.
- The research was published in a recent issue of the peer-reviewed journal Cell.
- The study documents how a paralysed individual used the BCI to control a robotic arm for seven consecutive months, with minimal calibration.
Key Scientific Challenge: Stability in BCIs
- A major hurdle in BCIs has been their instability—the brain’s signals change daily, making it hard for systems to consistently interpret intent.
- The research team studied neural patterns behind movement and how these shift over time.
Insight from Neuroscience
- Prof. Karunesh Ganguly, a UCSF neurologist, previously observed daily fluctuations in brain activity in animals and suspected similar changes occur in humans.
- These shifts in neural activity explained why earlier BCIs would lose effectiveness after a short time.
How does the System Work?
- The participant, who was paralysed due to a stroke, had tiny sensors implanted on his brain’s surface.
- These sensors did not stimulate the brain but instead read signals from areas related to movement intention.
Harnessing AI for Stability
- The participant imagined moving various body parts while the sensors captured his brain’s movement signals.
- Although physically immobile, his brain still generated distinct movement representations.
- Using AI and high-dimensional signal processing, researchers found that while the structure of signals remained stable, their location in the neural space shifted daily.
- A machine learning algorithm was designed to track and predict these shifts, allowing the BCI to maintain stability over several months.
From Training to Real-World Use
- Training began with a virtual robotic arm, which provided visual feedback to improve the participant’s control accuracy.
- The participant later transitioned to a real robotic arm, successfully performing tasks such as: Picking up, rotating, moving blocks, opening a cabinet, retrieving a cup, and using a water dispenser.
- These actions, though simple, are life-changing for individuals with severe paralysis.
Next Steps for BCI Technology
- According to Dr. Natraj, while the success is significant, further refinement is needed.
The goal is to make BCI systems fluid and reliable in complex real-world environments, such as navigating a crowded grocery store.
