2023-08-24 16:35:55
ALS and Stroke: AI Brings Language Back
San Francisco, Berkeley, Stanford, PTE, 24. August 2023
With the help of computer brain interfaces and artificial intelligence (AI), researchers from the University of California San Francisco (UCSF) and the University of California Berkeley have made a breakthrough in enabling individuals with neurodegenerative impairments to communicate once again. This groundbreaking research has given hope to patients suffering from conditions such as stroke and ALS, offering them the possibility to regain their ability to speak and express emotions.
One of the patients, Ann, had experienced a massive stroke that left her paralyzed and unable to use her muscles. Through the use of a digital avatar, Ann was able to convey her thoughts and emotions. Another patient, suffering from ALS, had their brain activity translated into text thanks to researchers at Stanford University.
The process involved the implantation of electrodes on the surface of the brain, precisely in the areas responsible for language production. These electrodes were then connected to a computer, allowing the researchers to capture the brain signals generated when the patients were mentally talking. By utilizing AI algorithms, these signals were translated into words, enabling the patients to communicate.
In order to ensure accuracy, the patients underwent extensive training programs to familiarize the AI system with their specific brain signals. Ann, for instance, learned 1,024 words, which the AI system could then interpret from her brain activity. Additionally, researchers developed an avatar that could mimic the patients’ facial expressions based on their thoughts, allowing them to express emotions such as joy, sadness, and surprise.
While the system has shown promising results, there is still room for improvement. The conversion of thoughts into spoken text currently has an average rate of 78 words per minute, compared to the natural conversation rate of approximately 160 words per minute among English speakers. Furthermore, the error rate stands at 25 percent, showcasing the need for further development.
The Stanford researchers employed a similar approach with their patient, Pat, who had two sensor arrays implanted on her brain’s surface. This allowed her to communicate at a rate of 62 words per minute using the AI system. However, it is essential to note that this is still a proof of concept and not a fully functional device for everyday use.
Dr. Francis Willett, a neurosurgeon from Stanford, emphasized the significance of this breakthrough, stating, “This is a big step forward in restoring rapid communication to people who are unable to speak due to paralysis.” While there is still work to be done, these advancements offer hope for individuals with neurodegenerative conditions, enabling them to regain vital communication abilities.]
#brings #language #Gütsel #Online #Gütersloh #OWL #live