2023-05-24 11:45:07
Scientists have created a non-invasive brain-computer interface that captures the user’s intended hand gestures in the user’s mind.
The achievement is the work of Yifeng Bu’s team, from the University of California at San Diego, United States.
The prototype created is a first step in the development of a non-invasive brain-computer interface that could one day allow patients with paralysis, limb amputation or other physical problems to use their minds to control an artificial device that helps them with daily tasks. .
What has been achieved so far in this line of research and development surpasses any previous achievement of this type.
For the new interface, the team has turned to magnetoencephalography (MEG). “MEG offers a safe and precise option to develop a brain-computer interface that could ultimately help patients,” says Mingxiong Huang, from the University of California, San Diego and a member of the research team.
The system uses a helmet equipped with 306 sensors to detect magnetic fields produced by neural electrical currents that circulate between neurons in the brain. Other alternative brain-computer interface techniques include electrocorticography (ECoG), which requires the surgical implantation of electrodes on the surface of the brain, and electroencephalography (EEG), which localizes brain activity non-invasively but with less precision.
The magnetoencephalography helmet used in the experiments of this study. (Illustration: MEG Center at UC San Diego Qualcomm Institute)
With magnetoencephalography, it is possible to pick up thoughts without drilling into the skull or placing electrodes in physical contact with the brain, notes Dr. Roland Lee of the University of California, San Diego and a member of the research team. Lee compares the safety of MEG to taking a patient’s temperature. “The MEG measures the magnetic energy that the brain emits, like a thermometer measures the heat that the body emits. That makes the MEG a completely safe and non-invasive technique.”
The authors of the new study tested the ability to use the MEG to distinguish between hand gestures performed by 12 volunteer subjects. These people were outfitted with the MEG helmet and randomly asked to perform one of the gestures used in the Rock, Paper, Scissors game (as in previous studies of this type).
When the results of the experiments were obtained, the researchers verified that their system made it possible to distinguish between hand gestures with an accuracy greater than 85%. These results are comparable to previous studies using an invasive brain-computer interface.
Bu and his colleagues report the technical details of their system and experiments with it in the academic journal Cerebral Cortex, under the title “Magnetoencephalogram-based brain–computer interface for hand-gesture decoding using deep learning.” (Fountain: NCYT de Amazings)
#read #brain #hand #gestures #thought #person