Publishing house Radiotekhnika

"Publishing house Radiotekhnika":
scientific and technical literature.
Books and journals of publishing houses: IPRZHR, RS-PRESS, SCIENCE-PRESS

Тел.: +7 (495) 625-9241


The Prediction Models of Coronary Heart Desease for Female Respondents on Basis of the Analysis of Psychological and Behavioural Qualities of the Person


O.V. Shatalova, E.A. Shashkova, A.B. Kraskovsky

The aim of work is the research of prediction effectivity of the coronary heart disease (CHD) of females on basis of use of new information technologies. The space of informative signs is produced on basis of the analysis of physiological and psychological features of the person. The parameters of arterial pressure and some biometric characteristics were used as physiological pre-diction parameters of CHD risk. The characteristics of accentuations received with the help of questionnaire of L.I. Sobchik were used as psychological prediction parameters of CHD risk. Correlation analysis was used for identification of dependence between prediction parameters in the chosen space of informative signs. Different structures of space of signs that had been used for discriminant analysis and neuronet modelling were received depending on correlation score of informative signs. It is shown as a research result that only the complete psychological description of the person but not separate psychological characteristics is influenced the propensity for CHD. The factorial analysis allowed to reduce the space of signs to six factors. The results of probability of the correct classification with the help of a neural network became a little worse (diagnostic responsivity (DR) = 97,47 %, diagnostic specificity (DS) = 98,04 %, diagnostic efficiency (DE) = 97,78 %) than the results (DR=100 %, DS=100 %, DE=100%) that are received with the help of neural network with the use of all psychological signs. However the usage of these six factors is better than the usage of the six the most significantly distinct signs (extroversion, aggression, rigidity, age, pulse, pulse pressure). The best classification from used methods is carried out with the help of neuronet modeling.
June 24, 2020
May 29, 2020

© Издательство «РАДИОТЕХНИКА», 2004-2017            Тел.: (495) 625-9241                   Designed by [SWAP]Studio