A Segmentation Of Mouse Organs Based On Atlas And BLT

Recently,with the rapid development, molecular imaging(MI) play an important role in the treatment of disease. This technology involved so many subjects and could observe disease and tumor in molecular and cells level. 3D optical tomography has gained extensive attentions due to its sensitivity, high safety, low cost and specificity. The way to get digital structure of small animals is an important part of 3D optical tomography. But there are some problems in existing methods.The research work of this thesis is the improvement of getting digital structure of small animals in 3D optical tomography. There are 3 problems in the existing methods of 3D optical tomography in our lab:(1) to segment Micro-CT data of small animals, the contrast agent is needed, which may cause the death of mouse in continuous observation;(2) Micro-CT data need segment manually, which cost too much time and energy;(3) Micro-CT system increase the complexity and cost of BLT system.To solve these problem, this thesis explore two solutions:First, to the optimization of segment Micro-CT data, this thesis show an BLT method based on atlas segmentation. This thesis use CT atlas of mouse to segment the Micro-CT data to get the structure information of mouse. And do the reconstruction of BLT. Through the experiment, it proves that the thesis can solve the problem that to segment Micro-CT data of small animals, the contrast agent is needed and Micro-CT data need segment manually, which cost too much time and energy.Then, to the optimization of gain 3D structure of mouse, this manuscript show a an BLT method based on atlas segmentation of structured lighting. This method through the structured lighting to get structured lighting data, segment structured lighting data by atlas segmentation. And do the reconstruction of BLT. Through the experiment, it proves that the thesis can solve all the problem that mentioned above. But the accuracy of result is need to improve.