*********************************
There is now a CONTENT FREEZE for Mercury while we switch to a new platform. It began on Friday, March 10 at 6pm and will end on Wednesday, March 15 at noon. No new content can be created during this time, but all material in the system as of the beginning of the freeze will be migrated to the new platform, including users and groups. Functionally the new site is identical to the old one. webteam@gatech.edu
*********************************
William Pilcher
BME PhD Thesis Proposal Presentation
Date: March 5th, 2020
Time: 12:00 PM
Location: Whitaker, McIntire Conference Room 3115
Committee Members:
Denis Tsygankov, PhD (Georgia Tech/Emory, Department of Biomedical Engineering) (Advisor)
Peng Qiu, PhD (Georgia Tech/Emory, Department of Biomedical Engineering)
May Wang, PhD (Georgia Tech/Emory, Department of Biomedical Engineering)
Shu Jia, PhD (Georgia Tech/Emory, Department of Biomedical Engineering)
Gregory Gibson (Georgia Tech, School of Biological Sciences)
Title: A Computational Platform for Efficient Characterization and Classification of Complex Shapes and Patterns in Biological Images
Summary: This project is designed to develop a new computational platform for comprehensive automated analysis of complex shapes and patterns in biomedical images. The tremendous progress in imaging technologies, such as fluorescent nanoparticle probes, super-resolution microscopy, and various optogenetic tools, provided biomedical researchers with new capabilities for precise manipulation and visualization of many essential processes in live cells and tissues. This progress, in turn, necessitates the advancement of quantitative approaches for automated image analysis. However, despite an increasing number and sophistication of such tools, the robust and unbiased extraction of descriptive features of cellular processes continues to be a bottleneck in many studies that involve generation of complex, highly variable, and content-rich imaging data. The overarching goal of the proposed study is to develop a broadly applicable morphometric approach that provides a mathematical mapping of arbitrarily complex shapes onto graph structures that, unlike traditional methods, preserve all the geometric information and allow for efficient and precise extraction and characterization of all structural features. Instead of relying on a large set of generic descriptors hoping that some of them are informative, in this method, a machine-learning algorithm receives a rich set of shape measures that are specific to the analyzed structure but extracted with a generic mapping. Thus, the project’s objective is to build a unique methodology, which is not context-specific but still powerful enough to automatically characterize subtle structural differences in imaging data. The data for thorough validation of the proposed methodology have been acquired through vascular tube formation experiments and also generated with a simulation model of collective endothelial cell behavior.
