Simultaneous Spectrophotometric Determination of Mefenamic Acid and Paracetamol in Pharmaceutical Preparation by Using Artificial Neural Network

Abstract

The present paper describes the spectrophotometric method for determination of mixture of paracetamol (PAR) and mefenamic acid (MEF) by using artificial neural network (ANN). The spectrum of PAR and MEF showed strong overlapping. Therefore, to reduce the overlapping, absorption data was extracted from the spectrum of the mixture of PAR and MEF and the concentration sets were used as input and output for ANN, respectively. A three layer feed-forward neural network with back-propagation algorithm was then constructed. To obtain the best topology for ANN, several parameters were optimized. Thus, in each step, root mean square error (RMSE) for both PAR and MEF was calculated. Finally, in the latter step, real sample based on conventional standard method (USP) was analyzed. The results obtained showed some differences between the standard method and the proposed method (ANN).