Publication of the journal

Izvestiya Vysshikh Uchebnykh Zavedenii. Elektronika

Publication of the journal

The section is currently being updated

New systems and algorithms for automating the control of drug administration into the human body require mandatory verification before testing directly on patients. The main task is to create a software tool capable of conducting a comprehensive evaluation of the efficiency of developed systems and algorithms of automated insulin therapy. In this work, a software concept for evaluating the efficiency of automated insulin therapy systems with three modes of operation is proposed. The three modes of operation are: testing the algorithm of insulin pump control, testing the mathematical model of human metabolism and model identification. The software is based on the mathematical model of human metabolism characterized by a high degree of completeness and accuracy. It was shown that numerical solution of the model for the given scenario of external perturbations and given control algorithm makes it possible to evaluate the control efficiency both graphically and by a set of statistical, technical and medical metrics. The simulation of errors in the technical elements of the insulin therapy system, such as insulin pump and glucose monitor, and of time and amplitude delays of user input of external perturbations data is implemented. It is possible to compare custom mathematical models, written as systems of ordinary differential equations, with the built-in model. The software that is open-source and written using Python language allows creation of a digital twin of a real patient for further use in simulations by identifying the built-in mathematical model from a set of experimental data. A graphical interface has been developed and methods for solving differential equations and optimization have been selected. The application of the developed software will allow the acceleration and simplification of the development process of new methods, control algorithms and mathematical models used to automate insulin therapy of patients with type 1 diabetes.
Dmitry A. Chuprakov
National Research University of Electronic Technology (Russia, 124498, Moscow, Zelenograd, Shokin sq., 1)
Kirill V. Pozhar
National Research University of Electronic Technology (Russia, 124498, Moscow, Zelenograd, Shokin sq., 1)
Elina I. Strukova
National Research University of Electronic Technology (Russia, 124498, Moscow, Zelenograd, Shokin sq., 1)
Nikita S. Kiryutochkin
National Research University of Electronic Technology (Russia, 124498, Moscow, Zelenograd, Shokin sq., 1)
© 2025. National Research University of Electronic Technology

124498, Moscow, Zelenograd, Bld. 1, Shokin Square, MIET, editorial office of the Journal "Proceedings of Universities. Electronics", room 7231

+7 (499) 734-62-05
magazine@miee.ru