Three-dimensional diffuse optical tomography: System implementation and validation of reconstruction algorithms

We present the development of an experimental three-dimensional (3-D) diffuse optical tomographic image reconstruction system for imaging of highly scattering tissue-mimicking material using near infrared laser light and the validation of 3-D reconstruction algorithms. The algorithm reconstructs the spatial distribution of the optical parameters of a volume from trans-illumination measurements on the boundary of the object under study. The emphasis is on system development and validation of 3-D model based iterative image reconstruction (MOBIIR) algorithms. We test the performance of the method for a cylindrical phantom with embedded absorbing perturbation. The system is based on frequency domain approach. The amplitude and phase measurements of modulated laser light exiting the object at multiple detector planes are carried out. We examine a 3-D finite element algorithm with cylindrical mesh to provide forward diffusion model to reconstruct 3-D and 2-D images of the interior absorption coefficient. The experimental photon intensity data in different planes have been compared with 3-D predicted measurement data. A good match has been observed between experimental measurements and the model prediction of light transport in tissue.