Application of Design of Experiments and Multilayer Perceptrons Neural Network in the Optimization of Diclofenac Sodium Extended Release Tablets with Carbopol (R) 71G

Abstract

The purpose of the study was to screen the effects of formulation factors on the in vitro release profile of diclofenac sodium from matrix tablets using design of experiment (DOE). Formulations of diclofenac sodium tablets, with Carbopol (R) 71G as matrix substance, were optimized by artificial neural network. According to Central Composite Design, 10 formulations of diclofenac sodium matrix tablets were prepared. As network inputs, concentration of Carbopol (R) 71G and the Kollidon (R) K-25 were selected. In vitro dissolution time profiles at 5 different sampling times were chosen as responses. The independent variables and the release parameters were processed by multilayer perceptrons neural network (MLP). Results of drug release studies indicate that drug release rates vary between different formulations, with a range of 1 h to more than 8 h to complete dissolution. For two tested formulations there was no difference between experimental and MLP predicted in vitro profiles. The M LP model was optimized. The root mean square value for the trained network was 0.07%, which indicated that the optimal MLP model was reached. The optimal tablet formulation predicted by MLP was with 23% of Carbopol (R) 710 and 0.8% of Kollidon (R) K-25. Calculated difference factor (f(1) 7.37) and similarity factor (f(2) 70.79) indicate that there is no difference between predicted and experimentally observed drug release profiles for the optimal formulation. The satisfactory prediction of drug release for optimal formulation by the MLP in this study has shown the applicability of this optimization method in modeling extended release tablet formulation.