The neurointerface made it possible to communicate with "locked" people

Oleg Lischuk, N+1

Participant of the study with a neurointerface (Wyss Center)

An international group of scientists has developed a system that allows patients with locked-in syndrome to answer "yes" or "no" to the questions asked. The results are published in the journal PLoS Biology (Chaudhary et al., Brain–Computer Interface–Based Communication in the Completely Locked-In State).

The "locked-in person" syndrome is a state of complete muscle paralysis in which the patient cannot move and talk, while being fully conscious with unaffected emotions and thinking. Some of these people may react to what is said with eye movements, the condition of the rest looks practically no different from a coma (in this case it is called the "completely locked person" syndrome). The cause of this condition can be head injuries, stroke, encephalitis, polio and various neurological disorders associated with damage to motor neurons or their axons.

The existing experimental neurocomputer interfaces (NCIS) can partially restore the ability of "locked" people to communicate, but require implantation of electrodes directly into the brain. Attempts to create similar interfaces based on the registration of brain waves (electroencephalography, EEG) and some other manifestations of brain activity have not yet been crowned with success.

In search of a non-invasive technology for communication with the "locked person" syndrome, the staff of the University of Tübingen with colleagues from other research centers in Germany, the USA, China and Switzerland have created a system based on near-infrared functional spectroscopy (fBICS, fNIRS). This method allows us to evaluate the activity of various parts of the cerebral cortex by the intensity of blood flow in them – hemoglobin, unlike tissues, absorbs near infrared radiation well.

Four patients took part in the tests of the system, in whom the "locked man" syndrome developed as a result of amyotrophic lateral sclerosis – an incurable disease in which the death of motor neurons leads to progressive muscle paralysis. During the experiment, the researchers used machine learning with a teacher (using the support vector method) to train an algorithm to distinguish between "yes" and "no" answers by changes in blood flow in the frontal-central areas of the cerebral cortex. To do this, they recorded such changes using the FBIX method at a time when patients reacted to obviously correct (for example, "Paris is the capital of France") or incorrect (for example, "Paris is the capital of Germany") statements. Three patients underwent 46 training sessions within a few weeks, the fourth – 20. In addition to fBICS, scientists recorded EEGs to check the participants' attentiveness – low-frequency delta and theta waves corresponding to sleep, nap or meditation correlated with an increase in the number of system errors.

As a result of the training, the system was able to distinguish between "yes" and "no" answers with 70 percent accuracy. After that, patients were asked questions, the answers to which are unknown (for example, "Do you have any pain?" or "Do you want to go to London?"). Each question was repeated 10 times, and if the system registered the answer "yes" in at least seven cases, it was considered the real opinion of the patient.

Thus, with the help of the system, scientists and relatives were able to communicate with "locked" patients. For example, one of the participants reported that she would like to see New York, another - to visit her brother in Spain, and an elderly patient expressed a negative attitude to the marriage of his granddaughter with a man younger than her.

The researchers were interested in the question of how satisfied their wards were with life in their condition, and, according to one of the authors of the work, Niels Birbaumer, all participants reported that "life is beautiful." These counterintuitive data correspond to the results of a 2011 study, according to which the majority of patients with stable locked-in syndrome are quite satisfied with life, which must be taken into account when processing requests for euthanasia (with the stabilization of the "locked-in" state, the attitude towards life may change for the better). As explained by another member of the research team Ujwal Chaudhary, there is a hypothesis that the brain of people in this state, trying to protect the mind, loses the ability to process negative emotions.

Groups of scientists around the world are engaged in the development and improvement of neurointerfaces, and many have managed to achieve considerable success. Examples include systems that allow you to control individual fingers of the prosthesis, independently eat, move and touch an artificial arm with limb paralysis, as well as partially restore the spinal cord after injury. Recently, an American-Chinese team managed to bring the speed of a non-invasive neurointerface to one character per second.

Portal "Eternal youth" http://vechnayamolodost.ru  02.02.2017