Application of a multivariate technique to Raman spectra for quantification of body chemicals.

Raman spectroscopy is a highly specific technique for the identification of molecules by way of the associated characteristic spectra. The aim of this feasibility study is to assess the combination of the multivariate calibration technique of Partial Least-Squares with Raman spectroscopy for the estimation of glucose, lactic acid, and urea concentrations in the presence of each other in a water substrate. The instrument is a CCD-based Raman spectrometer utilizing the 514.5 nm argon laser line. The estimates for the analyte concentrations yielded a standard deviation of concentration residuals of 20.71 mg/dL for glucose, 12.92 mg/dL for lactic acid, and 19.07 mg/dL for urea.