Net Analyte Signal Calculation in Near-Infrared Spectra Calibration

In this paper, net analyte signal (NAS) calculation was used for near-infrared (NIR) spectra calibration and prediction glucose concentration in urine sample. Near-infrared spectra in the 10000~4000 cm^-1 region were measured for human urine glucose with various concentrations at 37degC. The most informative spectral ranges of 4200~4800 cm^-1 were selected by principal component analysis (PCA) for glucose in the mixtures. For glucose, the correlation coefficient (R) of 0.9988 and the root mean square error of prediction (RMSEP) of 20.03 mg/dl were obtained by NAS and partial least squares (PLS). The NAS is the part of the spectrum that is directly related to the analyte of interest, and it can be used as a rapid and reliable tool for quantitative analysis of glucose in biological sample.