Optimization of diclofenac sodium dissolution from sustained release formulations using an artificial neural network

Abstract

The application of a two-level back propagation type of neural network has been demonstrated for studying and optimization of diclofenac sodium dissolution from sustained release matrix tablets. The effects of three formulation components (cetyl alcohol, polyvinylpyrrolidone K-30 and magnesium stearate) on the dissolution rate were analyzed using this method. A non-linear relationship between the amount of cetyl alcohol and PVP K-30 and the amount of dissolved drug is described. The convenience of a formulation study by two- and three-dimensional response surface analysis is presented. Neural network technique can be particularly suitable in the pharmaceutical technology of sustained release dosage forms where systems are complex and nonlinear relationships between independent and dependent variables often exist.