dislike of wearing the device. Thus, sensors are good but expensive, not affordable for everybody and could be uncomfortable. Therefore, the constant search for alternative solutions remains an important challenge. The purpose of the paper is to show the possibility of using hierarchical modeling technology to develop and study glycemic profile prediction algorithm as, to some extent, alternative to continuous monitoring sensors in a context of limited irregular measurements. Results. The program-algorithmic structure for realization of the concept of hierarchical simulation is developed. The possibility of conducting research on models of varying complexity is shown. An algorithm for insulin-glucose tolerance test was synthesized. A procedure for predicting the daily glycemic profile by analytical formulas has been developed, which provides an opportunity to assess the trend of glycemic dynamics as an addition to the irregular glucose measurements with a glucometer. A simulation study, the result of which is the visualization of glycemic profile in a context of expected food intake schedule and compensating insulin doses obtained by the analytical algorithm, was conducted. Conclusions. The proposed hierarchical modeling technology, based on the use of mathematical models of varying complexity, allows to conduct a complex of simulation studies to correct glycemia in diabetes at the preclinical and pre-ambulatory stages. During the simulation of forecasting procedure, configuration discrepancies of the glycemic profile obtained from different models were detected, but they are within the margin of error and reproduce the main trend in the dynamics of glycemia during meals and insulin injections. The calculated bolus doses of insulin are almost identical to those used by insulin-dependent patients. The simplicity of calculations using analytical formulas can be a prerequisite for the implementation of the algorithm in a special-purpose portable autonomous devices or in applications for Android OS.

Background. In recent years, modern technical devices have been created so that to use in the practice of treating diabetes mellitus. These are systems for continuous monitoring of glycemia, which is a significant addition to the widely accepted measurements of glucose levels with a glucometer, various infusion systems, which significantly improve the doctor's decision-making process. However, such technical means are quite expensive and inaccessible to a wide range of users. In addition, their use is associated with both adverse reactions when wearing them and with patient compliance issues. In this case an alternative can be using mathematical modeling tools. Objective. The aim of the paper is to prove the possibility of using mathematical modeling to predict the glycemic profile as a certain degree of alternative to a sensor for continuous monitoring of blood glucose levels under conditions of limited irregular measurements. Methods. To solve the problem it is proposed to employ the technology of mathematical modeling. The structure of the model makes it possible to implement the mathematical formalism by analytical formulae. Results. As a result, the insulin-glucose-tolerance test has been developed that allows quantitatively assessing a patient's personal sensitivity to insulin-bolus therapy. We proposed the mathematical model for solving the problem by analytical formulae. Algorithms for identifying model parameters, an algorithm for calculating the insulin dose that compensates for the carbohydrate component in the intended meal, and an algorithm for predicting the daily glycemic profile have been developed. The software-algorithmic structure for the implementation of the mathematical formalism has been developed. Conclusions. The conducted simulation study employing the technology of mathematical modeling makes it possible to evaluate the functioning of the developed procedures at the preclinical stage. The simplicity of calculations using analytical formulae can be a prerequisite for the implementation of the algorithm in portable autonomous special-purpose devices or in smartdata under the Android OS, which is a definite contribution to development of digital diabetology.