A new research report revealed a way to visualize a person’s brain activity as real-life images that reproduce what they observe in real time. The study was conducted by researchers from the Russian company Neurobotics and the Moscow Institute of Physics and Technology (MIPT), and according to them it will allow the development of new stroke-controlled stroke rehabilitation devices by brain signals.
The team of researchers has published the research in preview in the repository archive of preprints dedicated to biological sciences BioRxiv and posted a video showing his system of “mental reading” in activity.
According to the MIPT website, which reported on the research published in pre-print last month, to develop brain-controlled devices and methods for treating patients’ cognitive disorders and post-stroke rehabilitation, neurobiologists must understand how the brain codes information. Researchers do this by studying the brain activity of subjects who perceive visual information, for example, by watching a video.
The research team of MIPT and Neurobotics is not the first; solutions for the extraction of images observed from brain signals already exist. According to the research, existing solutions use either functional magnetic resonance imaging (MRI) or synchronous brain-computer interfaces (based on event-related potential recording) to make discrete selections.
However, according to the researchers, these two scenarios have advantages that, unfortunately, are outweighed by serious limitations that hinder the implementation of BCI technology in clinical practice and in everyday life, with applications requiring rapid and rapid control. real time of an external device being made impractical, inter alia, by these approaches.
According to the research report, one solution might be to combine the benefits of both scenarios. The brain-computer interface solution developed by the MIPT and Neurobobotics researchers is based on artificial neural networks and electroencephalography, or EEG, a technique for recording brain waves via electrodes placed non-invasively on the scalp.
By analyzing brain activity, the system reconstructs in real time the images seen by a person undergoing an EEG, according to the study.
Vladimir Konyshev, who heads MIPT’s neurorobotics lab, said of the research:
“We are working on the Neuronet Assistive Technologies project of the National Technology Initiative, which focuses on the brain-computer interface that allows stroke patients to control an exoskeleton arm for neurorehabilitation, or to paralyzed patients to drive. an electric wheelchair, for example. The ultimate goal is to increase the accuracy of neuronal control in healthy people as well. ”
Source : research