Quantitative evaluation of the impact of variation of optical parameters on the estimation of blood hematocrit and oxygen saturation for dual-wavelength photoacoustics

Abstract

Photoacoustic (PA) spectroscopy is considered to be one of the most effective ways to measure the levels of hematocrit (H) and oxygenation saturation (SO2) of blood, which are essential for diagnosing blood-related illnesses. This simulation study aims to investigate the impact of individual optical parameters, i.e., optical absorption coefficient (μ a ), scattering coefficient (μ s ), and anisotropy factor (g), on the accuracy of this technique in estimating the blood properties. We first performed the Monte Carlo simulations, using realistic optical parameters, to obtain the fluence maps for various samples. The wavelengths of the incident light were chosen to be 532, 700, 1000, and 1064 nm. Thereafter, the k-Wave simulations were executed, incorporating those fluence maps to generate the PA signals. The blood properties were obtained using the PA signals. We introduced variations in μ a , μ s , and g ranging from −10% to +10%, −10% to +10%, and −5% to +1%, respectively, at 700 and 1000 nm wavelengths. One parameter, at both wavelengths, was changed at a time, keeping others fixed. Subsequently, we examined how accurately the blood parameters could be determined at physiological hematocrit levels. A 10% variation in μ a induces a 10% change in H estimation but no change in SO2 determination. Almost no change has been seen for μ s variation. However, a 5% (−5% to 0%) variation in the g factor resulted in approximately 160% and 115% changes in the PA signal amplitudes at 700 and 1000 nm, respectively, leading to ≈125% error in hematocrit estimation and ≈14% deviation in SO2 assessment when nominal SO2=70%. It is clear from this study that the scattering anisotropy factor is a very sensitive parameter and a small change in its value can result in large errors in the PA estimation of blood properties. In the future, in vitro experiments with pathological blood (inducing variation in the g parameter) will be performed, and accordingly, the accuracy of the PA technique in quantifying blood H and SO2 will be evaluated.