The System And Algorithm Study Of Three-dimensional Bioluminescencetomography

As an important branch of optical molecular imaging, bioluminescence tomography(BLT) can analyze and detect many biological processes, such as the growth and metastasis of tumor in vivo at cell and molecular level and the expression of specific genes. At present, there are only few BLT imaging system products in market and the Xenogen of American accounts for the major share. So, it is important to develop our own BLT system with independent intellectual property rights and high performance. This paper discusses the bioluminescence tomography reconstruction algorithm and the design for the hardware system. The mainly works are shown as below:1. Researched the optical transmission model in the biological tissue. This paper converted the radiation transfer equation into diffusion equation for simplification and used the finite element method(FEM) to solve the equation.2. Simplified the optical transmission in the air andthe lens system as a pinhole model, gained linear relationship between gray and light radiation power based on photoelectric effect, and obtained the energy relationship between image point and object point based on Lambert’s cosine law. Reconstructed the surface light irradiance based on images accepted from CCD by deducing the pinhole model, and verified the algorithm by phantom and tissue experiment.3. Researched the bioluminescence tomography reconstruction algorithm. This paper proposed an adaptive selection strategy of the interested region which can reduce the illness of reconstruction algorithm. Furthermore, it can avoid human interference and improve the stability of the reconstruction procedure. The model order reduction method was then used after the selection of interested region to maintain the positive definiteness of the coefficient matrix in the equation transformation, and finally the light source was reconstructed by combining the Tihonov regularization and the conjugate vector base algorithm.4. Designed and built the hardware and software system of BLT imaging instrument. Through the rotating platform, the camera is able to acquire eight luminescence images with different perspectives of the mouse to increase useful prior information for reconstruction. Phantom and mouse experiments have been done to verify the performance of the BLT imaging system. The center point errors are less than 1mm and 2mm in phantom and mouse experiments.