Research On Tomography Imaging System To Obtain A High-Resolution Atlas Of The Mouse Brain

In spite of the rapid development of science and technology, the brain is still one of thebiggest unsolved mysteries. It is well known that the function is determined by the struc-ture. Thus, the brain functions depend more on the aggregate of neurons or the neuronalnetwork, which is considered to be the basis for understanding brain functions and diseases.In order to observe the neuronal network of brain, we need a three-dimensional resolution of1μm and a centimeter-sized detection range, Nevertheless, the classical imaging methods(e.g. magnetic resonance and electron microscope) cannot resolve the contradiction betweenlarge specimen and high-resolution. Because of the technological limitations, our recogni-tions on neural network and connectivity in mammalian whole brains remains inadequate.This paper discusses a method, a technique and an instrument to detect the intricate three-dimensional structure in large specimen, which will be applied to obtain the high-resolutionstructural atlas of whole mouse brain.To fill the gap uncovered by most current imaging methods at the mesoscopic level,the paper establishes the method and technique, which are suitable for volumetric imag-ing centimeter-sized specimens at sub-micron level. Base on the existing section-imagingmethod, the dissertation resolved three key technique issues: introducing a classic epi-?uorescence light path for absorption-contrast imaging, which improves the image contrastand resolution, and moreover, significantly simplifies the alignment of the imaging systemand makes the special design on knife possible; designing a simple but effective diamondknife and mounting style to keep high performance in a long period; designing a tomo-graphic sectioning mode, it is carried out column by column in the same layer to providea short and constant immersion time for each layer, which avoids non-uniform deforma-tion of the brain structure. As solving these issues, the dissertation eliminates the problemof chatter, which is always an unresolved doubt to worldwide counterparts in the past tenyears. Based on these technological breakthroughs, the dissertation developed a Micro-Optical Sectioning Tomography (MOST) instrument, which can obtain 1-micron-section continuously and stably, image the section with a resolution of 0.71μm, and accuratelysynchronize the sectioning and imaging. Through these innovative methods and techniquesabove, the MOST system can acquire data continuously, automatically and rapidly, the datahas natural and accurate registration, and the data acquisition has high stability and highaccuracy.MOST performed a 242-hour uninterrupted data acquisition to obtain a Golgi-Coxstained microstructure atlas of a whole mouse brain. The average rate of acquisition is0.11μs/voxel, and the total amount of uncompressed volume data exceeds 8 TB, covering15,380 coronal sections with a pixel resolution of 0.33μm by 0.33μm by 1.0μm. Somestudies were carried out on the basis of whole mouse brain dataset, includes the image pre-processing, three dimensional reconstruction and neurite tracing. It is difficult to observethe mismatch in MOST dataset, and the dataset can correspond with the traditional atlas ofwhole mouse brain. According to the result of literature search, it is the first whole brainmicrostructure atlas at the neurite level from a single mouse in the world.This mesoscale neuroanatomical structural atlas of the whole mouse brain provides abasic experimental dataset for digitizing the mouse brain architecture and simulating thebrain functions. MOST system is potential to map the whole brain atlas of mouse modelswith neurologic disease, and to obtain the network of blood vessels and capillary in wholemouse brain. It is likely that in combination with new developments in the techniquesof specimen preparation (e.g. the transgenic animal models with multi-labeling), MOSTsystem will help us to obtain functional connection atlas of a ?uorescent mouse brain withhigh resolution. One the other hand, MOST system also can be apply to the high-resolutionthree-dimensional structure imaging of other biological specimens (e.g. insects, embryos,organs and plants), even industrial materials as well.