Quantifying tissue hemodynamics by NIRS versus DOT: global versus focal changes in cerebral hemodynamics

Abstract

Near infrared spectroscopy (NIRS) is used to quantify changes in oxy-hemoglobin (HbO) and deoxy-hemoglobin (Hb) concentrations in tissue. The analysis uses the modified Beer-Lambert law, which is generally valid for quantifying global concentration changes. We examine the errors that result from analyzing focal changes in HbO and Hb concentrations. We find that the measured focal change in HbO and Hb are linearly proportional to the actual focal changes but that the proportionally constants are different. Thus relative changes in HbO and Hb cannot, in general, be quantified. However, we show that under certain circumstances it is possible to quantify these relative changes. This builds the case for diffuse optical tomography (DOT) which in general should be able to quantify focal changes in HbO and Hb through the use of image reconstruction algorithms that deconvolve the photon diffusion point-spread-function. We demonstrate the differences between NIRS and DOT using a rat model of somatosensory stimulation.