Nonlinear inversion technique for absorption tomography of turbid media using spatially resolved backscattered light

This report presents a proposal of a new technique to estimate the cross-sectional absorption distribution of turbid media from backscattered light by solving a nonlinear inverse problem. After illuminating a beam of light on the surface of a turbid object and measuring the backscattered light as a function of distance from the light incident point, we divide the object into multiple virtual layers to estimate the absorption distribution. The path lengths of photon propagation in the respective layers are calculated using Monte Carlo simulation. The absorption coefficient of each virtual layer can be estimated from the backscattered intensity and the path length distribution in a depth direction. For solving this inverse problem, the linear calculation results are useful as initial solutions. Then the final solutions are obtained from iteration of the nonlinear calculation. Convergence into a unique solution and robustness of the solution against the measurement noise were confirmed. The effectiveness of the proposed technique was verified through simulation and measurement. By lateral scanning of a source–detector pair, we can reconstruct a cross-sectional image of the turbid medium to the depth to which the detected light reaches.