Determination of cholesterol using a novel magnetohydrodynamic acoustic-resonance near-IR (MARNIR) spectrometer

Near-IR spectrometric determination of minor constituents of biological systems is complicated by the fact that near-IR spectra of these materials vary in different chemical and physical environments. In such cases, wavelength selection methods and full-spectral techniques such as partial least-squares and principal component regression (which weight each wavelength in calibration) produce excess error because they must attempt to model both variations in major constituents and variations in the analyte. A magnetohydrodynamic acoustic-resonance near-IR (MARNIR) spectrometer can determine major constituents of biological materials noninvasively and nondestructively, leaving the near-IR spectrum of the analyte to be used quantitatively with less prediction error.