Engineers at the University of California (UCLA) have created an innovative, wearable, and non-invasive Brain-Computer Interface (BCI) system. This device uses Artificial Intelligence (AI) as an assistant to more accurately decipher human intentions and perform various tasks based on them, such as controlling a robotic arm or a computer cursor.
According to a study published in the journal Nature Machine Intelligence, this new interface is a significant milestone in the development of non-invasive BCI systems. The technology may be particularly beneficial for people with physical limitations—those who are paralyzed or suffer from neurological diseases.
How the New BCI System Works
The engineering team developed special algorithms to decipher Electroencephalography (EEG) signals. The EEG method records the electrical activity of the brain and identifies the user’s movement intentions from these signals.
The deciphered signals were then connected to an AI-based platform that interprets the user’s intentions in real-time. With the help of this AI system, people perform tasks much faster.
Overcoming Limitations of Existing Technology
Currently existing, surgically implanted BCI devices manage to convert brain signals into commands, but their use is limited due to the high risks and costs associated with neurosurgery. Although the first such devices appeared more than 20 years ago, they are still restricted to small pilot clinical trials.
To overcome these limitations, researchers created and tested a new, non-invasive BCI system aided by artificial intelligence. Four participants were involved in the experiment: three without motor impairments and one who was paralyzed from the waist down.
The participants wore an EEG helmet that recorded their brain’s electrical signals. Then, using special algorithms, the researchers converted these signals into movements for a computer cursor and a robotic arm. Simultaneously, the AI system, with a built-in camera, observed the deciphered movements in real-time and helped the participants complete two different tasks: controlling a computer cursor and a robotic arm.
Demonstrated Success
The experiment showed that all four participants completed the tasks significantly faster with the help of artificial intelligence.
It is particularly noteworthy that the paralyzed participant was only able to control the robotic arm with the assistance of the AI; without it, they could not complete the task.
