Scientists are exploring a number of ways for people with disabilities to communicate with their thoughts. Being a part of a larger study where the safety of implantable devices is evaluated, researchers from Stanford University have developed new software that decodes neural activity from a device implanted in a person’s brain while the person imagines writing letters and words by hand.
By attempting handwriting, the study participant with the device implanted in their brain, typed 90 characters per minute, realizing the fastest typing speed for any type of brain-computer interface with over 94% accuracy in real-time and over 99% accuracy when paired with autocorrect.
Though there are already a few existing BCIs such as EEG spellers, the typing speed is relatively slower and the accuracy is lower. Most of the existing BCIs are used for large movements such as reaching. With success in improving accuracy and speed, this new system can provide help for people who cannot speak due to paralysis.
Further research is needed for this new system to be applied in the real world as this study only provides proof that this approach is feasible. The BCI system still requires further improvements to perform at a faster speed and with less training time. However, this proposed device is realistic and will be able to help paralyzed individuals to communicate in the real world.
Why this matters: This study is a breakthrough in the use of machine learning to aid the life of people with impaired speech or motor skills. With this advancement in BCI technology, new implant devices may be developed to help restore communication in disabled patients.
Subject to further development, this innovation could let people with paralysis rapidly type without using their hands.