Separability and cross talk: optimizing dual wavelength combinations for near-infrared spectroscopy of the adult head

By means of noninvasive near-infrared spectroscopy (NIRS), cerebral concentration changes in oxygenated and deoxygenated hemoglobin ([oxy-Hb] and [deoxy-Hb]) can be determined. The quality of the concentration changes' assessment critically depends on the wavelength combination used. Trying to optimize this combination, two spectroscopic effects must be taken into account: cross talk and separability. Cross talk between [oxy-Hb] and [deoxy-Hb] occurs because the assumption made in the analysis-that there is a homogeneous concentration change-does not hold true for the adult human head. Separability-to be introduced in this paper-is a measure for the degree of physical noise of the measurement that will influence the noise of the concentration changes' assessment. In other words, high separability corresponds to a low noise with respect to the concentration changes assessed. Here, we present analytical expressions for both measures and provide model-based estimates of cross talk and separability for any combination of two wavelengths between 610 and 920 nm. These theoretical considerations allow for two predictions: (a) if both wavelengths used are greater than approximately 780 nm, cross talk is high and separability is low resulting in erroneous and noisy concentration data. (b) If one wavelength is chosen below 720 nm while the other is greater than 730 nm, cross talk is low and separability is high resulting in accurate concentration changes. We show the relevance of these theoretical results for noninvasive NIRS by testing the predictions on experimental data obtained in adults undergoing visual stimulation.