We use the optical/near-infrared (NIR) reflectance spectroscopy to non-invasively measure the hemoglobin saturation in living human skin. The difficulties in clinical application of this technique for skin tissue oxygenation monitoring are due to the complexity of extracting the information of chromophores distribution and their concentrations from the reflectance spectra in case of multiple scattering of light. We have developed a computational model of human skin and Monte Carlo technique for simulation of the reflectance spectra of skin in visible and near-infrared spectral region. The computational model of skin contains several layers with wavy inter-layered boundaries corresponding to the cell structure of human skin. Our model takes into account probe geometry, variations of spatial distribution of blood vessels, various levels of blood oxygen saturation, volume fraction of water, oxy- and deoxy-hemoglobin, melanin content and chromophores of interest. The small source-detector separation (250, 400 and 800 micrometers ) required due to the shallow (100-150 micrometers under skin surface) spatial location of skin capillary loops is of our main interest. Comparison of the results of spectra simulation and experimental results made in vivo are presented. As the experimental system we use the spectrometer mating with the two-dimensional array of CCD camera.© (2001) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.