Use of artificial neural networks to predict drug dissolution profiles andevaluation of network performance using similarity factor

Purpose. To use artificial neural networks for predicting dissolution profi les of matrix-controlled release theophylline pellet preparation, and to ev aluate the network performance by comparing the predicted dissolution profi les with those obtained from physical experiments using similarity factor. Methods. The Multi-Layered Perceptron (MLP) neural network was used to pred ict the dissolution profiles of theophylline pellets containing different r atios of microcrystalline cellulose (MCC) and glyceryl monostearate (GMS). The concepts of leave-one-out as well as a time-point by time-point estimat ion basis were used to predict the rate of drug release for each matrix rat io. All the data were used for training, except for one set which was selec ted to compare with the predicted output. The closeness between the predict ed and the reference dissolution profiles was investigated using similarity factor (f(2)). Results. The f(2) values were all above 60, indicating that the predicted d issolution profiles were closely similar to the dissolution profiles obtain ed from physical experiments. Conclusion. The MLP network could be used as a model for predicting the dis solution profiles of matrix-controlled release theophylline pellet preparat ion in product development.