Have you ever wondered what it would be like to communicate with your thoughts alone? Imagine being able to express yourself without speaking or writing, just by thinking of the words you want to say. This may sound like science fiction, but it could soon become a reality for some people who have lost their ability to speak due to neurological conditions.
A team of researchers at Duke University has developed a brain implant that can decode signals from the brain’s speech center and predict the sounds a person is trying to articulate. The device, which consists of 256 microscopic brain sensors on a flexible plastic sheet, was tested on four patients undergoing brain surgery. The patients repeated nonsensical words while the device recorded their brain activity. The device then used a machine learning algorithm to predict speech sounds based on brain activity. The results are promising as certain sounds were predicted with up to 84 percent accuracy, and an average of 40 percent accuracy for 90-second data samples.
The researchers are working on improving the prediction accuracy and developing wireless versions of the device. Their ultimate goal is to enable natural speech communication for people with motor disorders that impair their speech, such as ALS, Parkinson’s disease, MS, cerebral palsy, dystonia, and stroke. These disorders affect the body’s ability to control movement, and can result from damage or disease affecting the nervous system. The current tools available to help these people communicate are generally slow and cumbersome, such as eye-tracking devices or keyboards.
The brain implant by Duke University researchers has the potential of revolutionizing the field of adaptive technology and improve the quality of life for individuals with motor disorders. (Have you thought of AI yet?) The device could also open new possibilities for human-computer interaction and brain-computer interfaces. The study is published in the journal Nature Communications.