Abstract
In this study, two chemometric techniques, partial least-square regression (PLSR) and artificial neural network (ANN) were developed and compared for the simultaneous assay of paracetamol (PCT) and caffeine (CAF) in pharmaceutical formulations by using spectrophotometric data. Six different concentrations of paracetamol and caffeine were considered to make mixture solutions of standard samples by using orthogonal experimental design (OED). UV spectra of these mixtures were recorded in the wavelength range of 205-300 nm versus a solvent blank and digitized absorbance was sampled at 1 nm intervals. Drug concentrations and instrumental spectra of 36 mixture solutions were used for model development and validation and finally 6 commercially available tablets were used to test the developed models. ANN shows better prediction efficiencies than that of PLSR with R2 value 99.28% for prediction and 99.13% for validation set. These two models were successfully applied to commercial pharmaceutical formulations, and it is found by ANN that the drugs contain 75 to 86% of paracetamol and 77 to 92% of caffeine of their label claim. Either of the proposed methods is simple and rapid and can be easily used in the quality assessment of drugs as an alternative analytical method.
 Bangladesh J. Sci. Ind. Res.54(3), 215-222, 2019
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call