In Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy

Pavel Govyadinov

will defend his dissertation

Segmentation, Analysis and Visualization of Large Microvascular Networks in the Mouse Brain

Abstract

Advances in high-throughput imaging allow researchers to collect three-dimensional images of whole organ microvascular networks. These networks are highly complex, making them difficult to segment, visualize, analyze, and synthesize. These limitations make it particularly difficult to study tissue microstructure, since microvasculature plays a prominent role in tissue development and disease progression. The underlying complexity stems from microvascular networks tending to be highly interconnected and heterogeneous. This presents a difficult visualization problem, since visual clarity is disrupted by the high density. This structure also hampers analysis, making it difficult to study microvascular changes as a consequence of disease progression. Current research suggests that microvascular networks play a particularly prominent role in neurodegenerative disease. In this dissertation, I present a framework for working with microvascular networks embedded in multi-terabyte, three-dimensional images collected using high-throughput microscopy. I propose an efficient and tuneable algorithm for segmentation utilizing graphics hardware, making this framework accessible to researchers relying on limited hardware in standard workstations. I propose multiple visualization methods that provide scientists with the ability to highlight important features in complex microvasculature structures useful for researchers of disease, as well as lay a foundation for a data structure for studying neurodegenerative disease. I show how this graph-based data structure is used for collecting statistical features from segmented networks, and how the metrics can be used to classify differences in networks due to tissue type or disease progression.

Date: Monday, July 22, 2019
Time: 1:00 PM - 2:00 PM
Place: Engineering 2 Bldg., W309
Advisor: Dr. Guoning Chen

Faculty, students, and the general public are invited.