Inverse method 3-D reconstruction of localized in vivo fluorescence-application to Sjogren syndrome

The development of specific fluorescently labeled cell surface markers have opened the possibility of specific and quantitative noninvasive diagnosis of tissue changes. We are developing a fluorescence scanning imaging system that can perform a "noninvasive optical biopsy" of the Sjogren syndrome (SS) which may replace the currently used histological biopsy. The diagnosis of SS is based on the quantification of the number of topical preadministered fluorescent antibodies which specifically bind to the lymphocytes infiltrating the minor salivary glands. We intend to scan the lower lip, and for each position of the scan, generate a two-dimensional (2-D) image of fluorescence using a charge-coupled device (CCD) camera. We have shown previously that our diffuse fluorescent photon migration theory predicts adequately the positions and strengths of one and two fluorescent targets embedded at different depths in tissue-like phantoms. An inverse reconstruction algorithm based on our theoretical findings has been written in C/sup ++/ and uses 2-D images to predict the strength and location of embedded fluorophores. However, due to large numbers of variables, which include the optical properties of the tissue at the excitation and emission wavelengths, and the positions and strengths of an unknown number of fluorophore targets, the validity of the final result depends on assumptions (such as the number of targets) and the input values for the optical parameters. Our results show that the number of fluorophore targets reconstructed for each scan is limited to two, and at least the scattering coefficient at the excitation wavelength is needed a priori to obtain good results. The latter can be obtained by measurements of spatially resolved diffuse reflectance at the excitation wavelength that provides the product of the absorption and scattering coefficients.