Premium
Score‐based quantitative principal component analysis with application to the study of active pharmaceutical ingredients based on attenuated total reflection fourier‐transform‐infrared spectra
Author(s) -
BalcerowskaCzerniak Grażyna,
Kupcewicz Bogumiła
Publication year - 2017
Publication title -
journal of chemometrics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.47
H-Index - 92
eISSN - 1099-128X
pISSN - 0886-9383
DOI - 10.1002/cem.2863
Subject(s) - principal component analysis , principal component regression , partial least squares regression , attenuated total reflection , chemometrics , analyte , fourier transform , reflection (computer programming) , calibration , interference (communication) , least squares function approximation , biological system , spectral line , mathematics , chemistry , analytical chemistry (journal) , statistics , pattern recognition (psychology) , infrared spectroscopy , computer science , artificial intelligence , chromatography , mathematical analysis , computer network , channel (broadcasting) , organic chemistry , estimator , biology , programming language , physics , astronomy
This paper presents a novel algorithm, named Score‐based Quantitative Principal Component Analysis, for the quantification of different active ingredients in pharmaceutical tablets based on attenuated total reflection Fourier‐transform‐infrared spectra. The procedure is proposed for solving the problem of the simultaneous determination of the amount of several analytes in the presence of unknown interfering compounds. This is accomplished by incorporating into the analysis a Principal Component Analysis calibration data set model obtained for 2‐way mixture spectra which are deemed as a linear combination of all pure component spectral signals of known amount of the analyte of interest. A relatively simple model involving the relationship between principal component (PC)‐scores for mixture spectrum and PC‐scores for pure component spectra has been developed for predicting the amount of a compound in the presence of an unknown interference in a new sample. The performance of the proposed method has been evaluated using 2 commonly used measures of accuracy, and it was also compared with 3 classical methods, ordinary least squares, partial least squares regression and multivariate curve resolution‐alternating least squares. The results show that in multicomponent 2‐way system spectra in cases where classical methods are not able to predict the constituent concentration accurately the PC‐scores based approach, under some circumstances, can generate good predictive results. The power of the proposed method has been evaluated by applying it to the analysis of simulated and experimental data.