Open AccessDevelopment of a methodological approach to determine regression equations in the study of the technology for manufacturing tablets based on quercetin
Author(s) -
О. В. Кутова,
І. V. Kovalevska,
Nataliya Demchenko
Publication year - 2022
Publication title -
vìsnik farmacìï
Language(s) - English
Resource type - Journals
eISSN - 2415-8844
pISSN - 1562-7241
DOI - 10.24959/nphj.22.79
Subject(s) - computer science , regression analysis , identification (biology) , regression , mathematics , data mining , statistics , machine learning , botany , biology
When developing the technology of tablets and tablet masses one of the tasks is to determine the amounts of excipients required to obtain pharmaco-technological quality indicators that meet the requirements of the State Pharmacopoeiaof Ukraine (SPhU).Aim. To develop an algorithm for determining the type of three-factor mathematical models with dependent variables.Materials and methods. The study object was experimental observations of the quantitative effect of excipients in the tablet composition based on a solid dispersion of quercetin on the pharmacopeial characteristics of this dosage form, in particular on its flowability. The experimental data were processed by the planned experiment using Mathcad 15 and MS Excel software according to the algorithm proposed.Results and discussion. It has been found that the identification of mathematical models in pharmaceutical studies with three dependent factors, which total value is determined by the quantitative composition of the dosage formand fixed at a certain level, is difficult to perform due to the difficulty of interpreting the multiple regression parameters as characteristics of factors in isolation through their correlability. It has been proven that the replacement of variablesleads to the determination of a mathematical model that does not reveal the mechanism of the factors’ action and is a static description of their overall impact on the indicator studied.Conclusions. As a result of the research conducted, regression equations were found to determine the effect of the amount of these excipients on the tablet flowability. It has been found that the most influential role in determining the target in the factor space studied is played by the interaction of factors. The algorithm for determining the mathematical description of the dependence with three variables has been proposed. Based on the fact that the research is conditioned by strict conditions in the quantitative composition, the chosen mathematical model of the dependence of the quadratic equation with two factors, for which there are limitations, has no correlation. The model determined is characterized by the possibility of providing the graphical interpretation and simplification of analysis and can be used for forecasting and optimization.