Three-Dimensional Quantification of Cerebral Microvasculature and Cerebral Microhemorrhages
- Xie, Danny Fei
- Advisor(s): Choi, Bernard
Abstract
Histology is the study of the microscopic anatomy of cells and tissue. The thickness of tissue sections in histology is limited by the ability of light to penetrate through tissue samples.Greater light penetration can be achieved through tissue clearing, the process of using chemical solvents to increase transparency in tissue samples by reducing refractive index mismatch. Tissue clearing enables the imaging of larger tissue samples and whole organs. This can be paired with confocal microscopy or other similar modalities to achieve three-dimensional (3D) imaging. Imaging large volumes of tissue in 3D can provide structural and morphological information that is not available with standard histology. This is particularly advantageous for studies involving structures that span across a large region and cannot be reliably captured within a single planar view, such as the microvascular network. The primary function of the microvascular network is to transport substances to and from tissue. This task is vital to the health and function of all organs throughout the body and is particularly important for the brain which receives about 20% of cardiac output despite being only 2% of the body by mass. Impairments to the brain vasculature can severely impact brain health and lead to cognitive impairment. When studying brain-related diseases, it is important to consider the vascular component of the brain. Visualizing the brain vasculature can be performed with standard histology but this is unable to fully capture the vascular network as it spans across multiple directions. Tissue clearing can enable 3D visualization of the microvascular network in the brain and other organs. In addition to 3D visualization, quantitative analysis is needed to facilitate studies of the microvascular network. Automated procedures are essential to enable timely and objective analysis. Accessible methods for quantitative analysis can help increase the feasibility of researchers to conduct microvascular-related studies. An indicator of impairment to the brain vasculature is cerebral microbleeds. Cerebral microbleeds are the result of chronic accumulation of blood in the brain. Cerebral microhemorrhages (CMH) are the precursor of these microbleeds. The presence of CMH increases with age and is often associated with increased risks for cognitive impairment and strokes. The mechanisms of CMH formation remain unclear, and one question in particular is what type of blood vessel in the brain is the likely source of CMH. To address this gap in knowledge, 3D visualization of the brain microvasculature and CMH is needed. We first developed a simple method for visualization of the 3D cerebral microvasculature. This procedure used an exogenous perfusion-based vascular label, tissue clearing with organic solvents, and confocal imaging to obtain 3D images of the brain microvasculature. The contrast-to-background with this method was greater than that of an endogenous vascular label. Next, we developed tools to facilitate automated analysis of 3D images of brain vasculature, specifically focusing on segmentation and measurement of diameter. The automated analysis was validated by comparing it to a manual analysis, with the manual analysis representing the ground truth. We found good agreement between automated results and manual results. We then applied these tools to study the relationship between CMH and the cerebral microvasculature. We specifically focused on identifying candidate blood vessels that may be the source of CMH. Vessel diameters were used to classify what type of vessels were nearest to a CMH lesion. We found that capillary-sized vessels are the likely vascular source of CMH. Lastly, we applied 3D histology to the characterization and quantification of CMH. We found novel information about the dimensions and morphology of CMH that was previously unknown using standard histology. Characterization of CMH with 3D histology suggests that some variability may occur in CMH quantification with 2D histology due to sampling and chance. Collectively, these findings highlight the advantages of using 3D histology to address problems that cannot be captured by 2D histology. The tools developed here to facilitate 3D histology of brain vasculature can be used to identify therapeutic targets to prevent impairment of brain health and function.