Quantitative Reconstruction of Absorption Coefficients for Photoacoustic Tomography

Abstract

Photoacoustic (PA) tomography (PAT) is a cutting-edge imaging modality for visualizing the internal structure and light-absorption distribution in tissue. However, reconstruction of the absorption distribution has been limited by nonuniform light fluence. This paper introduces a novel method for quantitative reconstruction of the distribution of optical absorption coefficients in tissue. In this method, we implement an iterative algorithm for recovering absorption coefficients from optical absorbed energy maps based on a 3D Monte Carlo simulation of light transport and integrated with fluence compensation to obtain the initialization parameters. In the iteration algorithm, we calculate the deviation between the detected and the computed absorbed energy distribution at each iteration. By minimizing the deviation in the absorbed energy, the recovered values converge to the true absorption distribution. The results of numerical simulation and phantom experiment theoretically and experimentally demonstrate that the proposed method performs an accurately quantitative estimate of the distribution of optical absorption coefficients. This work expects to provide accurate quantitative information for absorbers within tissues or organs, and thereby broaden the clinical applications of PAT.

Observing temporal variation in hemolysis through photoacoustics with a low cost LASER diode based system

“…brain tissue). An iterative algorithm has been also proposed to recover absorption coefficients from optical absorbed energy maps based on a 3D Monte Carlo simulation of light transport 45 . Similar approaches can be adopted in future study of the PA technique to estimate hemolysis in-vivo.…”

Deep learning acceleration of iterative model-based light fluence correction for photoacoustic tomography

“…This leads to inaccuracy in the simulation and correction of LF. To solve this problem, model-based iterative LF correction algorithm is proposed [17] , [18] , [19] , [20] , [21] . This method does not require the analytic form of the model, instead, it continually updates the unknown parameters through iterative optimization until the output of the solver matches the measured data [9] .…”

“…Liu et al . [20] applied an iterative algorithm based on 3D Monte Carlo simulation of light transport to achieve 3D light correction in PAT, but it was limited to simple numerical simulation and phantom experiments. In order to improve the correction accuracy and simplify the calculation process, Zhang et al .…”