2007 年 127 巻 10 号 p. 1723-1729
The estimation of paracetamol and orphenadrine citrate in a multicomponent pharmaceutical dosage form by spectrophotometric method has been reported. Because of highly interference in the spectra and the presence of non-linearity caused by the analyte concentrations which deviate from Beer and Lambert's law, partial least-squares (PLS) and artificial neural networks (ANN) techniques were used for the calibration. A validation set of spiked samples was employed for testing the accuracy and precision of the methods. Reasonably good recoveries were obtained with PLS for paracetamol and the use of an ANN allowed the estimation of orphenadrine citrate, a minor component which could not be adequately modeled by PLS. Three production batches of a commercial sample were analysed, and there was statistically no significant difference (P<0.05) between the results with the proposed method and those obtain with the official comparative method.