An algorithm for localizing inhomogeneities in pulsed diffuse optical tomography is proposed and implemented. A distinctive feature of this technique is the formation of an initial approximation to...

A method of diffuse optical tomography (DOT) is described for the direct detection of absorbing and scattering inhomogeneities such as cysts, hematoma, or tumors. DOT can also be used to detect small changes in the oxygenation and deoxygenation of blood in designated regions of the heads of neonates, and in breast cancer. In contrast to the early part of time-resolved transmittance of a femtosecond pulse, late-arriving photons are used for this purpose. A 2D numerical simulation is used for quantitative matching of diffuse model curves to experimental data. The finite difference method is used for computer simulation of movements of a cloud of photons from the surface of an object to its center. The 3D representations of the time point spread functions (TPSF) represent a flat surface in a homogeneous case. Lower parts of the surface indicate the presence of inhomogeneities at the corresponding angles.

Features of the propagation of low-coherent pulse radiation in phantoms of soft biological tissues (blood-perfused tissues of brain, breast, etc.) are described. Results of mathematical simulation of photon migration based on the diffusion approximation to the radiative transfer equation and Monte Carlo statistical testing are compared. It is corroborated that the photon density normalized maximum (PDNM) in relatively homogeneous strongly scattering media moves towards the center of the object under study. In the presence of inhomogeneities, the character of the PDNM motion changes; the absorbing inhomogeneity in a strongly scattering medium shifts the PDNM motion path towards the point symmetric to the center of this inhomogeneity with respect to the center of the object under study. In cases with a scattering inhomogeneity, the PDNM moves towards its center.