| Many non-invasive imaging techniques based upon hemodynamic responses of blood vessels have provided data for analysis of brain vasculature. However, these data lack the detail that could be ascertained by exploiting state-of-the-art neuroinformatics tools. A more complete and sufficiently detailed analysis of brain vascular architecture is critical for a wide variety of applications, including the development of models which may help predict or treat cerebrovascular disease.;Taking advantage of image stacks acquired from 3T time-of-flight magnetic resonance angiography and techniques previously used to create 3D neuronal reconstructions, the circle of Willis and the six major arteries that stem from it were reconstructed for sixty-one healthy subjects. The basis of this dissertation research is that recently available neuroinformatics tools can be exploited to create new models of vascular reconstructions that are representative of the general population of healthy subjects. These models can be used to generate more detailed novel comparative assessments between normal and diseased vasculature found in forms of cerebrovascular disease.;Sixty-one digital reconstructions of healthy human brain vasculature were created and extensive quantitative morphometrical analysis was conducted in order to characterize the anatomy of the human brain vessels, on both global and local levels of size, distances, angles and topology, concentrating in particular on vessel bifurcations. Additional analysis was conducted to provide quantitative description of the arterial branches, bifurcation patterns, shape and geographical distribution of the main cerebral arterial arborizations, as well as estimations of the corresponding vascular territories. Also as part of this analysis, computational models were created to examine fluid dynamics. In order to create these models, the brain was segmented and normal blood flow and wall shear stress values were measured, with the intent to provide a set of baseline values that could then be compared to patient data values. Lastly, the digital reconstructions and their extracted morphological measurements were archived in a database that will be made publicly available.;The information that can be extracted from these detailed reconstructions can be used to examine questions such as how the brain vasculature differs across various populations, such as males and females, different age groups, and healthy individuals versus individuals with cerebrovascular disease. Also, the information that can be extracted from the reconstructions can be used to implement complex computational models of fluid dynamics that may aid in the development of new treatments for individuals afflicted with cerebrovascular disease. Finally, the new information generated here may play a fundamental role in our ability to recognize vascular defects that could help treat, or even predict, cerebrovascular diseases. |
