A new approach to the development of methods for personalized expert analysis of laboratory data

Abstract

The presented work is devoted to the development of new approaches to the individual expert analysis of the obtained values  of laboratory data. A description of the sequence of the stages of mathematical transformations, the creation of a» panel «  neural network, the formation of matrix tables. Referring to their earlier publications, the authors argue that the structural  changes in the «panels» of the ratios of indicators formed by the rows of «reference points», despite the same type of displacement of the absolute parameters of the selected value, could in different observations both coincide and differ significantly,  while demonstrating the selective relationships that are justified in the known literature data. According to the authors, the  proposed algorithm makes it possible to establish hidden links between the dynamics of various laboratory indicators in individual cases, thereby significantly increasing their informativeness. The development and implementation of this method of  analysis, firstly, will allow to identify in the individual laboratory data, at least as «Express» method, States (images) corresponding to different complex pathological changes, including those requiring for its diagnosis of labor-intensive and expensive  research, and secondly-significantly expand the information content of routine laboratory research and without additional labor  and financial costs to be used in any health facilities.