Net analyte signal and radial basis function neural network for development spectrophotometry method for the simultaneous determination of metformin and sitagliptin in anti-diabetic commercial tablet

Abstract

In this study, the net analyte signal (NAS) and radial basis function neural network (RBF-NN) are presented for the simultaneous spectrophotometric analysis of metformin (MET) and sitagliptin (STG) in mixtures and tablet formulation. The coefficient of determination (R2) related to the test set of NAS procedure were achieved 0.9942 and 0.9853 for MET and STG, respectively. Also, mean recovery and root mean square error (RMSE) were found 99.66, 99.24% and 0.105, 1.116 for MET and STG, respectively. In the RBF-NN model, mean square error (MSE) with 23 epochs was 1.25 × 10−22 and 1.07 × 10−21 for MET and STG, respectively. Relative standard deviation (RSD) value was less than 1.5 in the analysis of the anti-diabetic tablet (Zipmet) for both approaches. The statistical comparison using analysis of variance (ANOVA) test proved that there was no significant difference between the proposed methods and the high-performance liquid chromatography (HPLC).