Objectives

The purpose of the research is the automatic processing and analysis of high-amount (nearly 1500) and high-resolution synchrotron based tomography images of mammalian cerebral cortical tissues. The objective was to acquire quantitative data allowing future blood flow modeling based on the structure of cerebral vessel network. . The brain samples were processed with histological method compatible with synchrotron x-ray microtomography. The method developed performs automated extraction of the relevant parameters of the vessel network on a standard personal computer using previously set parameters.

Method

The method implemented on the image series performs automatic segmentation by using determined intensity values. The procedure finds the boundary of the vessels and provides quantitative measures of the segments, and it also creates the topology of the vessel network.

Results

Our method produces topologically correct network of continuous, branching micro blood vessels of the sample studied. We present multiple options for the visualization of the detected vessels. The data structure also contains the statistics of each detected cross-section, in addition with the numerical quality statistics of the continuous vessel branches (e. g. vessel length or mean diameter). The algorithm resulted in highly accurate fitting at micrometer geometric resolution.

Conclusion

The reconstructed vessel network is the geometrically accurate description of the blood vessels in the tissue. The data acquired could be the base of a blood vessel surface reconstruction, and by adding physiological information it could be used for simulations in normal and pathological conditions (e. g. arteriosclerosis).  The precise vascular geometry and the distance distribution provides information about the barely understood neurovascular coupling, which is the basis of such modern diagnostic and research tools as the fMRI. After analyzing samples from multiple species or brain regions the difference could be determined objectively.

DOI: 10.17489/biohun/2013/1/01