Methodological Investigations On In Vivo Acquisition Of Target Optical Structures For Fluorescence Molecular Tomography

Fluorescence molecular tomography(FMT) has been emerging as a promising tool for small-animal imaging. The modality achieves the in vivo visualization of biological processes at cellular and molecular levels by reconstructing the three-dimensional(3D) distribution of the fluorescent targets without invasive investigation. In view of low imaging resolution and quantitative accuracy in FMT which are caused by the assumption of homogeneous optical structural background, a investigation on obtaining the heterogeneous background optical structures is proposed for improving the sensitivity of FMT. In our study, we choose the diffuse optical tomography(DOT) to obtaining the background optical structures.Relying on the deep penetration of scattering light in tissue, DOT is a effective noninvasive optical imaging technolgoy with lowcost for obtaining the tissue optical structrues. However, duo to strong scattering nature of most tissue types and relatively limited number of the available measurements, the conventional voxel-based DOT image reconstruction is a severely ill-posed inverse problem, leading to a region-based DOT is developed for in vivo reconstructing background optical structures with the aid of anatomical priors from X-ray CT and/or MRI anatomical imaging modalities. The method is based on the framework of the voxel-based DOT methodology and on an asssumption that different anatomical regions have their respective sets of homogeneous optical property distributions, combining with the photon transmission model of inverse solving strategy, for developing a effective region-based DOT under continous-wave(CW) mode. The simultaneously reconstructed optical backgrounds are then used directly in the FMT to improving the sensitivity of image reconstruction for evaluating the effectiveness of this approach.According to the different handling strategies of the background structural a priori information, the structural a priori method can be broadly classified into the hard-prior(HP) regularization, i.e., the region-based DOT method, and the soft-prior(SP) regularization. Specially, the HP regularization exhibits an decent performance in the quantitative accuracy when the structural priors is nearly accurate, while its quantitative performance is highly depended on the accuracy of the background structrual priors. Comparatively, the SP method is robust and unbiased in the presence of uncertainty in structural priors. Therefore, we present a quantitative comparison between the HP- and SP-DOT methods(the structural-based DOT) for acquiring the heterogeneous optical backgrounds with some imperfect structural priors to improving the sensitivity of FMT reconstruction.Although, with the aid of some imperfect anatomical a priori information, the structural-based DOT with either the HP or SP scheme holds promise for in vivo acquiring the optical background of tissue, the low robustness of the HP scheme to the segmentation error and the inferior performance of the SP one in the quantitative accuracy limit its further application. Therefore, we propose a novel shape-based DOT method for not only effectively determining the regional optical properties by the HP regularization but potentially achieving reasonable structural amelioration based on the spherical harmonics parameterization of the interior regions to improving the accuracy of the optical properties reconstruction by the HP regularization, leading itself to FMT for comparably improved recovery of fluorescence distribution to validate the effectiveness of the shape-based DOT method for obtaining the optical structures of the heterogeneous backgound.These proposed schemes above are not only numerical validated using a region-labeled 3D digital mouse model, which is developed from the CT images and cryosection data, and also conducted using several phantom experiments to assess the effectiveness of these methods for improving the sensitivity of FMT.