Studies On Reconstruction Algorithm For Optical Tomography

Optical tomography(OT) is an imaging modality with the near infrared light as sources for the optical parameters such as the scattering and absorption coefficients of tissues. Owing to the fact that the functional information of tissues is directly related to the optical parameters reconstructed, OT is a functional imaging modality for biomedical purpose. Because of its harmlessness as well as other superiorities such as its characteristics of real time and portability, which are impossible for other imaging modalities, this technique has attracted much attention and is of great value for research.In this thesis, the accuracy and the validity conditions of the diffusion equation as a forward model for the optical tomography was firstly studied. Simulation results for the forward projection of OT are presented for the cases of strong and low scattering media respectively based on Monte-Carlo (MC) method along with some discussions. Experiment results show that the diffusion equation can be used to describe the transportation process of near infrared photons accurately in scattering media. However, a relatively large error exists for the case of low scattering tissue, and as a result, this model can not be employed for the OT reconstruction in this case. Then, a new regularized conjugate-gradient reconstruction algorithm was proposed for optical tomograpy, in which the ill-posedness of the reconstruction problem and as a result the quality of the reconstructed image are improved by introducing two regularization terms of image entropy and the local smoothing function. Finally, a multi-resolution reconstruction algorithm was proposed for optical tomography. In this algorithm, the Delaunay triangulation rules are first used in the generation of the triangular mesh with...