We investigated a quantitative imaging of reduced scattering coefficients μs'( λ) and the absorption coefficients μa( λ) of in vivo cortical tissues in the range from visible to near-infrared (NIR) wavelengths based on diffuse reflectance spectral imaging technique. In this method, diffuse reflectance images of in vivo cortical tissue are acquired at nine wavelengths (500, 520, 540, 560, 570, 580, 600, 730, and 760 nm). A multiple regression analysis aided by the Monte Carlo simulation for the absorbance spectra is then utilized to estimate the optical coefficients of cortical tissue. This analysis calculates the concentration of oxygenated hemoglobin and that of deoxygenated hemoglobin, the scattering amplitude a and the scattering power b. The spectrum of absorption coefficient is deduced from the estimated concentrations of oxygenated hemoglobin and deoxygenated hemoglobin. The spectrum of reduced scattering coefficient is determined by the estimated scattering amplitude and scattering power. The particle size distribution of microstructure is calculated from the estimated scattering power b for evaluating the morphological change in brain tissue quantitatively. Animal experiments with in vivo exposed brain of rats demonstrated that the responses of the absorption properties to hyperoxic and anoxic conditions are in agreement with the expected well-known cortical hemodynamics. The average particle size was significantly reduced immediately after the onset of anoxia and then it was changed into an increase, which implied the swelling and shrinkage of the cellular and subcellular structures induced by loss of tissue viability in brain tissue.
Keywords: Monte Carlo simulation; Multispectral imaging; absorption coefficient; hemodynamics; particle sizing; reduced scattering coefficient; tissue morphology.