Rapid simultaneous quantitative determination of linagliptin and empagliflozin as antidiabetic drugs using spectrophotometric method based on fuzzy systems and radial basis function neural network in tablet formulation and biological sample

Abstract

In this study, spectrophotometry method was combined with artificial intelligence techniques for developing a simple, rapid, low-cost, and accurate approach for the simultaneous determination of Linagliptin (LIN) and Empagliflozin (EMPA) as antidiabetic drugs in mixtures and pharmaceutical formulation. Fuzzy inference system (FIS), adaptive Neuro-fuzzy inference system (ANFIS), and radial basis function neural network (RBF-NN) were proposed to predict the concentration of pharmaceutical components. The mean recovery and root mean square error (RMSE) of FIS were 98.44%, 103.50% and 0.0668, 0.1982 for LIN and EMPA, respectively. The mean recovery percentage of 100.07 and 99.83, as well as RMSE of 0.0033 and 0.0176 were obtained for LIN and EMPA, respectively. On the other hand, MSE related to the RBF-NN was found 1.46 × 10−26 and 1.72 × 10−25 for LIN and EMPA, respectively. Analysis of variance (ANOVA) test did not show a significant difference between high-performance liquid chromatography (HPLC) and proposed methods. The mean recovery (between 91.40% and 98.29%) and relative standard deviation (RSD) (lower than 0.5%) revealed that the suggested methods in a complex matrix such as urine had good predictive ability. Due to the obtained results, these methods can be used in quality control.