Analysis of Methods for Calculating Optimal Parameters for Insulin Boluses in Automated Insulin Therapy Systems with Control Based on Predictive Models.

Methods and algorithms suitable for solving stiff systems of nonlinear differential equations used in mathematical modeling of blood glucose dynamics are discussed, along with multivariate optimization algorithms for finding optimal values for insulin bolus parameters. The results show that use of a...

Descripción completa

Detalles Bibliográficos
Publicado en:Biomedical Engineering Vol. 57; no. 2; pp. 102 - 107
Autores principales: Chuprakov, D. A., Pozhar, K. V.
Formato: research tables/charts Journal Article
Publicado: Springer Nature Jul2023
Acceso en línea:Ver este registro en EBSCOhost
Descripción
Sumario:Methods and algorithms suitable for solving stiff systems of nonlinear differential equations used in mathematical modeling of blood glucose dynamics are discussed, along with multivariate optimization algorithms for finding optimal values for insulin bolus parameters. The results show that use of an Adams/BDF hybrid algorithm (LSODA) allows the systems of equations of the mathematical model to be solved with maximal efficiency, while the COBYLA optimization method provides the greatest accuracy and speed of finding the maximum of the efficiency function. These methods can be used in automated insulin therapy systems with control based on predictive models for calculating insulin boluses.