Australian researchers have developed an artificial intelligence (AI) model that can decipher words and sentences from brainwaves detected by a wearable cap, opening up a new frontier in medical science.
It will not only help patients with cognition, but could also revolutionise how humans interact with computers, the Australian Broadcasting Corporation reported on Monday.
The pioneering AI model, developed by expert Daniel Leong, PhD student Charles Zhou, and his supervisor Chin-Teng Lin at the GrapheneX-Human-centric Artificial Intelligence Center, University of Technology Sydney, used deep learning to translate the brain signals from electroencephalogram (EEG) into specific words.
Lin noted that the AI model at this stage learned from a limited collection of words and sentences to make it easier to detect individual words.
90% accuracy target
Researchers are recruiting more people to read the texts while wearing the EEG cap to refine the model.
They also intend to use it for communication between two people.
According to Mohit Shivdasani, a bioelectronics expert at the University of New South Wales, researchers have been looking for the patterns in biological signals "forever," but now AI can recognise brainwave patterns that have never been identified previously.
AI, particularly when used in implantable devices, could quickly personalise the brainwaves to how an individual completes a task, he added.
The team has so far achieved about 75 percent accuracy in converting thoughts to text, while aiming for 90 percent perfection.
