Sulavko, A. E., Lozhnikov, P. S., Choban, A. G., Stadnikov, D. G., Nigrey, A. A., Inivatov, D. P. (2020). Evaluation of EEG identification potential using statistical approach and convolutional neural networks. Information and Control Systems, (6), 37-49. https://doi.org/10.31799/1684-8853-2020-6-37-49

Introduction: Electroencephalograms contain information about the individual characteristics of the brain activities and the psychophysiological state of a subject. Purpose: To evaluate the identification potential of EEG, and to develop methods for the identification of users, their psychophysiological states and activities performed on a computer by their EEGs using convolutional neural networks.
Results: The information content of EEG rhythms was assessed from the viewpoint of the possibility to identify a person and his/her state. A high accuracy of determining the identity (98.5–99.99% for 10 electrodes, 96.47% for two electrodes Fp1 and Fp2) with a low transit time (2–2.5 s) was achieved. A significant decrease in accuracy was detected if the person was in different psychophysiological states during the training and testing. In earlier studies, this aspect was not given enough attention.
A method is proposed for increasing the robustness of personality recognition in altered psychophysiological states. An accuracy of 82–94% was achieved in recognizing states of alcohol intoxication, drowsiness or physical fatigue, and of 77.8–98.72% in recognizing the user’s activities (reading, typing or watching video).
Practical relevance: The results can be applied in security and remote monitoring applications

The article discusses the relationship of the keyboard handwriting of a computer user and the parameters of his electroencephalogram (EEG). As part of the work, an experiment was conducted to collect EEG data of 65 subjects who entered the passphrase on different keyboards at different times. An EEG analysis was performed, patterns and EEG parameters (features) were identified that can be used for person biometric identification. A method for identifying a person by the characteristics of the EEG in the process of keyboard input is proposed. A computational experiment was conducted with a large volume of test sample to assess the reliability of recognition of subjects. According to the results of the experiment, 1.62% errors were obtained. At the same time, no dependence of the EEG signs on the keyboard used by the subjects and the time of day, as well as on the variability of the signs with time, was detected.

Static biometric patterns such as fingerprint, iris and face are difficult to keep secret. Since the open pattern has a little potential replacement options stealing a strange open biometrics provides great opportunities for compromising systems. Authentication on the basis of electroencephalogram pattern (EEG) is the most secure kind of biometric security. The present study aims to develop a method of biometric authentication by the EEG data with high accuracy. Several neural network EEG pattern verification algorithms have been tested. A method for verification of the human EEG pattern based on a modified Bayes hypothesis formula has been developed. The following error indicators FAR <10-4 with FRR = 0.062 were achieved.

This paper is devoted to the problem of biometric identification of a person using the pa-rameters of electroencephalogram (EEG), recorded when observing visual stimuli. The principles of the human visual system are described, based on which an experiment was developed and conducted to collect EEG data of subjects for further analysis in order to create methods of biometric identification and authentication of the person. The ad-vantages of the developed methods are described, due to the high complexity of falsifying biometric EEG images. 65 subjects participated in the collection of biometric data. Two types of visual stimuli were used. All subjects underwent both series of experiments at least twice on different days. The differences of EEG signals from different people, which are manifested in large statistics, are revealed. A method is proposed for identifying a per-son by EEG parameters, based on the use of the Bayes hypothesis formula. The results of the experiment are presented. The following error rates were achieved - less than 0.001% with an EEG sample size of 7.5 seconds (not a single error was recorded for 53000 exper-iments), and also 0.54% for an EEG sample size of 2.5 seconds

The development of methods and technologies for the automatic determination of the psychophysiological state (PPS) of a person is an actual scientific and technical task. Early detection of the fact that the subject is in a sleepy state or in a state of intoxication at the workplace will help to
avoid accidents, harm to life, health, and causing losses. In this work the EEG data of 30 subjects in normal, sleepy conditions and a state of mild intoxication were collected. As a result of the spectral and correlation analysis of the EEG data features were selected. An amount of information about the
difference of the investigated states contained in the features was determined. A computational experiment on the recognition of human state according to EEG data based on the "naive" Bayes classifier was conducted. The following error level was achieved: 10.9% when recognizing the state of
“norm” and “intoxication”; 0.2% when recognizing the status of "normal" and "falling asleep."

Formulation of the problem.The psychophysiological state (PPS) of a person directly affects his ability to work.This work examines the following PFS: stress, sleep (all phases and stages), drowsiness (falling asleep), and alcohol intoxication.These conditions are the most important from the point of view of the need for their timely identification during the professional activities of employees whose work is associated with a high concentration of attention and increased danger. The purpose of the work is an analytical and synthetic study of the problem of automatic assessment (recognition) of a person’s psychophysiological state based on electroencephalogram (EEG) parameters. Results.Existing methods and approaches to determining these states based on EEG parameters are analyzed, described, and generalized.The results achieved in identifying these states using machine learning and pattern recognition methods are presented. Practical significance.Most of the reviewed studies rely on the analysis of EEG rhythms that correlate with MCP.We can highlight the most promising methods for recognizing states using neural networks and cascades of classifiers.Despite the high level of accuracy of the approaches considered, the methods tested in laboratory conditions have a number of problems that are currently unresolved and require further research.