Optical tomography in the presence of void regions

There is a growing interest in the use of near-infrared spectroscopy for th e noninvasive determination of the oxygenation level within biological tiss ue. Stemming from this application, there has been further research in the use of this technique for obtaining tomographic images of the neonatal head , with the view of determining the levels of oxygenated and deoxygenated bl ood within the brain. Owing to computational complexity, methods used for n umerical modeling of photon transfer within tissue have usually been limite d to the diffusion approximation of the Boltzmann transport equation. The d iffusion approximation, however, is not valid in regions of low scatter, su ch as the cerebrospinal fluid. Methods have been proposed for dealing with nonscattering regions within diffusing materials through the use of a radio sity-diffusion model. Currently, this new model assumes prior knowledge of the void region location; therefore it is instructive to examine the errors introduced in applying a simple diffusion-based reconstruction scheme in c ases in which there exists a nonscattering region. We present reconstructed images of objects that contain a nonscattering region within a diffusive m aterial. Here the forward data is calculated with the radiosity-diffusion m odel, and the inverse problem is solved with either the radiosity-diffusion model or the diffusion-only model. The reconstructed images show that even in the presence of only a thin nonscattering layer, a diffusion-only recon struction will fail. When a radiosity-diffusion model is used for image rec onstruction, together with a priori information about the position of the n onscattering region, the quality of the reconstructed image is considerably improved. The accuracy of the reconstructed images depends largely on the position of the anomaly with respect to the nonscattering region as well as the thickness of the nonscattering region. (C) 2000 Optical Society of Ame rica [S0740-3232(00)00809-7] OCIS codes: 170.3660, 290.1990, 100.3010.