ABSTRACTS OF ARTICLES OF THE JOURNAL "INFORMATION TECHNOLOGIES".
No. 9. Vol. 26. 2020

DOI: 10.17587/it.26.515-522

N. N. Yakhno^1,2, Professor, e-mail: info@ditc.ras.ru, V. N. Gridin¹, Professor, e-mail: info@ditc.ras.ru, D. S. Smirnof¹, Senior Researcher, e-mail: info@ditc.ras.ru, V. S. Panishchev¹, Senior Researcher, e-mail: info@ditc.ras.ru, V. A. Parfenov², Professor, e-mail: info@ditc.ras.ru, T. M. Ostroumova^1,2, Junior Researcher, e-mail: info@ditc.ras.ru, N. N. Koberskaya^1,2, Senior Researcher, e-mail: info@ditc.ras.ru,
¹ Design Information Technologies Center Russian Academy of Sciences, Odintsovo, Moscow Region, Russian Federation
² I. M. Sechenov First Medical State University, Moscow

Statistical Processing and Methods for Reducing the Dimension of Space on the Example of Data for Patients in the Analysis of Cognitive Impairment

The article proposes the development of methods for statistical processing of information and algorithms for converting and analyzing medical data of patients in order to detect factors of subjective and easy cognitive decline. The inapplicability of classical classification models for such data with a low initial consistency is shown. A technique is proposed for identifying groups of indicators that have the greatest consistency and dividing ability for a priori groups of patients. The data transformed in accordance with the methodology can be the initial data for constructing effective predictive models of the regression type.
Keyworlds: cognitive decline, spatial dimensionality reduction, multidimensional statistical methods, logistic regression, principal component analysis

P. 515–522

To the contents