ПОБУДОВА МОДЕЛЕЙ-КЛАСИФІКАТОРІВ ЗА РЕЗУЛЬТАТАМИ БАГАТОВИМІРНОЇ ПОЛЯРИЗАЦІЙНОЇ МІКРОСКОПІЇ В ТЕХНОЛОГІЇ СУДОВО-МЕДИЧНОГО ІНТЕЛЕКТУАЛЬНОГО МОНІТОРИНГУ ХВОРОБ СЕРЦЯ

Abstract

The significant prevalence of the cases of acute coronary insufficiency in the practice of a foren-sic expert and its suddenness gives a rise to suspicion of forensic investigating authorities about the vio-lent nature of the death of a person. It requires the use of objective and precise methods for diagnosis of acute ischemia of the myocardium. In this work the results of application of the methodology of creation of information technologies of multilevel intelligent monitoring are presented in order to provide data of de-cision-making processes by forensic medical expert. The methods of multidimensional polarization mi-croscopy and statistical processing of data with methods of inductive modeling are combined for con-structing a methodology for the creation of intelligent systems of multilevel forensic medical monitoring. On the example of the process of posthumous diagnosis of coronary heart disease and acute coronary in-sufficiency, processes of coordination of interactions of various types of formation of an array of informa-tive features, typical aggregates of synthesis of model-classifiers at each of the stages of monitoring are researched. To obtain informative features, a model of biological tissue of the myocardium was developed and the main diagnostic parameters were determined (statistical moments of 1–4 orders of coordinate distributions of the values of azimuths and the ellipticity of polarization and their autocorrelation func-tions, as well as wavelet coefficients of the corresponding distributions), which are dynamic due to its ne-crotic changes. Numerical characteristics of informative features form an array of input data for the syn-thesizer of models of monitoring information system. The classification of these data was provided by con-structing a decisive rule for the synthesizer on a multi-path algorithm of GMDH. The efficiency of the de-scribed methodology has been experimentally proved.