Key Takeaways
- Researchers at UCSF have developed a brain implant that allows people with paralysis to speak and gesture at the same time.
- The implant, tested on two participants, achieved 84% accuracy in predicting speech and 88% in predicting gestures.
- This breakthrough could significantly enhance communication for those with paralysis.
Scientists at the University of California, San Francisco (UCSF) have made a significant advancement in brain-computer interface technology, enabling individuals with paralysis to communicate more naturally and effectively.
The study, led by Samantha Brosler, involved two participants who had lost almost all movement in their limbs due to paralysis from a stroke and amyotrophic lateral sclerosis (ALS).
The team recorded the participants' brain activity while they attempted to say 10 phrases and perform 10 gestures, separately and simultaneously, to train custom machine learning models.
These models were then used to predict the participants' intended phrases and gestures, which were displayed on a screen and controlled by an avatar.
The results were promising, with the models achieving 84% accuracy in predicting speech and 88% in predicting gestures for one participant, and 66% and 70% accuracy for the other.
Brosler explained, 'In our everyday life, gestures play an important role in communication. Saying 'maybe' and nodding your head has a very different connotation than if you shake your head.'
The use of avatars was chosen to simplify the proof of concept, as it allowed the researchers to demonstrate that both actions could be decoded from brain activity.
This technology could revolutionize communication for those with paralysis, offering a more nuanced and natural way to interact with the world.
The findings were published in a New Scientist article, highlighting the potential of this technology to improve the quality of life for individuals with paralysis.
In our everyday life, gestures play an important role in communication.
Samantha Brosler, Researcher at UCSF
