Font Size: a A A

Bioluminescence Tomography Imaging Modality Fusion And Light Source Localization Method Research

Posted on:2015-03-31Degree:MasterType:Thesis
Country:ChinaCandidate:N N NingFull Text:PDF
GTID:2308330482985204Subject:Pattern Recognition and Intelligent Systems
Abstract/Summary:PDF Full Text Request
People can observe organism physiological activities in cell and molecular level by the emergence of molecular imaging. As one of optical molecular imaging representative modality, bioluminescence imaging has an advantage of high sensitivity, fast imaging and low cost, which has been widely applied in various fields of life science. Bioluminescence imaging is non-contact imaging, inner light source can be reverse reconstruction by biology three-dimensional surface bioluminescence flux distribution, thus reflecting the biological changes in the body from the cell and molecular level. However, its implementation process is very complex, including optical image acquisition,3D biological surface reconstruction,3D surface bioluminescence flux reconstruction and internal fluorescent light source reconstruction. Therefore, the reconstruction of the various steps method is important for bioluminescence tomography.The problem is what can we do to rebuild bioluminescence light simply, rapidly and accurately. In this paper we mainly studies from the following aspects:first, we improved three dimensional surface reconstruction algorithm based on voxel. The contradiction between subdivision level and the memory was solved by introducing level controlled error parameter E. At the same time, we set up an all-optical bioluminescence tomography system by combine with the improved algorithm and the needs of small animal bioluminescence imaging. Experimental results demonstrate the applicability of the proposed method. Second, we propose a BLT and micro-CT modality fusion method. In this method, the registration between small animal optical surface based on the multi-angle projection and micro-CT surface is realized by iterated closest points method. Then, combined with the improvement of texture mapping method, fluorescence simplified model in free space transmission and fluorescence quantitative correction to realize the fusion between BLT and micro-CT modality. This method can avoid the strict system settings and the random error by human interaction. The experimental results show that this method is of higher registration precision, and the fusion result can be applied to real small animal experiments of bioluminescence light reconstruction. Last, we propose a kind of all-optical bioluminescence reconstruction algorithm. In this method, first reconstruct small animal 3D surface by multi-angle optical projection. Then, combine fluorescence simplified model in free space transmission and fluorescence quantitative correction to reconstruct the 3D fluorescence luminous flux distributions. At the last, implement the small animal bioluminescence light source localization base on the finite element method to solve the simplified model of radiative transfer equation. Compared with the implement combine with micro-CT. this method can avoid ionizing radiation and tedious process step. Experimental results show that the method has good localization accuracy and can be preliminary application in small animals bioluminescence imaging in clinical trials.
Keywords/Search Tags:optical surface reconstruction, bioluminescence reconstruction, luminous flux reconstruction, fusion between BLT and micro-CT, iterated closest points
PDF Full Text Request
Related items