Pixel-wise reconstruction of tissue absorption coefficients in photoacoustic tomography using a non-segmentation iterative method.

In Photoacoustic Tomography (PAT), the acquired image represents a light energy deposition map of the imaging object. For quantitative imaging, the PAT image is converted into an absorption coefficient ( ) map by dividing the light fluence (LF). Previous methods usually assume a uniform tissue distribution, and consequently degrade the LF correction results. Here, we propose a simple method to reconstruct the pixel-wise map. Our method is based on a non-segmentation-based iterative algorithm, which alternately optimizes the LF distribution and the map. Using simulation data, as well as phantom and animal data, we implemented our algorithm and compared it to segmentation-based correction methods. The results show that our method can obtain accurate estimation of the LF distribution and therefore improve the image quality and feature visibility of the map. Our method may facilitate efficient calculation of the concentration distributions of endogenous and exogenous agents in vivo.