Learned spectral decoloring enables photoacoustic oximetry

Abstract

The ability of photoacoustic imaging to measure functional tissue properties, such as blood oxygenation sO_2, enables a wide variety of possible applications. sO_2 can be computed from the ratio of oxyhemoglobin HbO_2 and deoxyhemoglobin Hb, which can be distuinguished by multispectral photoacoustic imaging due to their distinct wavelength-dependent absorption. However, current methods for estimating sO_2 yield inaccurate results in realistic settings, due to the unknown and wavelength-dependent influence of the light fluence on the signal. In this work, we propose learned spectral decoloring to enable blood oxygenation measurements to be inferred from multispectral photoacoustic imaging. The method computes sO_2 pixel-wise, directly from initial pressure spectra S_{text {p}_0}(lambda , mathbf {x}), which represent initial pressure values at a fixed spatial location mathbf {x} over all recorded wavelengths lambda. The method is compared to linear unmixing approaches, as well as pO_2 and blood gas analysis reference measurements. Experimental results suggest that the proposed method is able to obtain sO_2 estimates from multispectral photoacoustic measurements in silico, in vitro, and in vivo.