Computer simulation of time-resolved optical imaging of objects hidden in turbid media

We review research on time-resolved optical imaging of objects hidden in strongly scattering media, with emphasis on the application to breast cancer detection. A method is presented to simulate the propagation of light in turbid media. Based on a numerical algorithm to solve the time-dependent diffusion equation, the method takes into account spatial variations of the reduced scattering and absorption factors of the medium due to the presence of objects as well as random fluctuations of these factors. It is shown that the simulation method reproduces, without fitting, experimental results on tissue-like phantoms. Using experimental and simulation results, an assessment is made of the reliability for extracting the reduced scattering and absorption coefficients of the medium from time-resolved reflection and transillumination data. The simulation technique is employed to study the conditions for locating mm-sized objects immersed in a turbid medium, by direct, time-resolved imaging. We discuss a simple method to enhance the imaging power of the time-resolved technique. The mathematical justification of the method, as well as some applications to simple problems, is given. The simulation technique is employed to demonstrate the effectiveness of the data processing technique. Results of time-resolved reflection experiments and simulations are presented, showing that the use of the latter allow us to locate 1 mm diameter objects under conditions which would prevent detection otherwise. Our results demonstrate that the combination of simulation and the appropriate processing of the diffusive part of the time-resolved reflected or transmitted light intensity may substantially increase the potential of the time-resolved near-infrared diffusive light imaging technique as a diagnostic tool for breast cancer detection. (C) 1998 Elsevier Science B.V. All rights reserved.