Date of Award
Spring 5-15-2018
Degree Name
Doctor of Philosophy (PhD)
Degree Type
Dissertation
Abstract
Optical imaging for medical applications is a growing field, and it has the potential to improve medical outcomes through its increased sensitivity and specificity, lower cost, and small instrumentation footprint as compared to other imaging modalities. The method holds great promise, ranging from direct clinical use as a diagnostic or therapeutic tool, to pre-clinical applications for increased understanding of pathology. Additionally, optical imaging uses non-ionizing radiation which is safe for patients, so it can be used for repeated imaging procedures to monitor therapy, guide treatment, and provide real-time feedback. The versatile features of fluorescence-based optical imaging make it suited for cancer related imaging applications to increase patient survival and improve clinical outcomes. This dissertation focuses on the development of image processing methods to obtain semi-quantitative fluorescence imaging data. These methods allow for the standardization of fluorescence imaging data for tumor characterization.
When a fluorophore is located within tissue, changes in the fluorescence intensity can be used to isolate structures of interest. Typically, this is done through the accumulation of a dye in a target tissue either by the enhanced permeation and retention effect (EPR), or through targeted peptide sequences that bind receptors present in specific tissue types. When imaged, the contrast generated by a fluorescent probe can be used to indicate the presence or absence of a structure, bio-chemical compound, or receptor. Fluorescence intensity contrast can answer many biological and clinical questions effectively; however, we were interested in analyzing more than solely contrast when using planar fluorescence imaging.
To better understand tumor properties, we developed a series of algorithms that harness additional pieces of information present in the fluorescence signal. We demonstrated that adding novel image processing algorithms enhanced the knowledge obtained from planar fluorescence images. Through this work, we gained an understanding of alternative approaches for processing planar fluorescence imaging data with the goal of improving future cancer diagnostics and therapeutics.
Language
English (en)
Chair
Samuel Achilefu
Committee Members
Rebecca Aft, Mark Anastasio, Jianmin Cui, Joseph Culver,
Comments
Permanent URL: https://doi.org/10.7936/K71V5DFV