Use of Geometric Methods of Analysis of Video-Oculographic Data to Assess the Functional Condition of a Person

Objective. In article describes two original algorithms for the analysis of video-oculographic data and analyses the effectiveness of these algorithms to assess the current functional condition of a person. One of the algorithms is designed for estimating macrosaccades curvature and the other - to evaluate the smoothness of target tracking. Both algorithms are based on geometric methods of videooculographic data processing. Methods. The assess of the algorithms effectiveness was realized on the model of alcohol intoxication (used the medium doses of alcohol - 0.8 g of 96 % alcohol per 1 kg of body weight). For the simulation of saccadic movements and smooth tracking we developed two psychomotor tests, which were evaluated by two key indicators: the curvature of microsaccade and smooth target tracking. Results. The results showed that the operator's activity disorders were usually accompanied by disturbance of the smooth oculomotor tracking. However the significant changes in the curvature of macrosaccades were not observed. Conclusions. Indicators of the smooth oculomotor tracking turned out to be quite informative for assessing the functional state of a person during the activity and can be used for practical diagnosis. Indicators of curvature of microsaccade were not sufficiently sensitive to the negative external factors and can't be used for practical diagnosis of the current condition of the person

Algorithm · Computer vision · Curvature · Machine learning · Smoothness · Advanced Scientific Research Methods · Computer Science · Mathematics · Psychology · Technology and Human Factors in Education and Health · Artificial Intelligence