Date of Award

December 2016

Degree Type

Dissertation

Degree Name

Doctor of Philosophy

Department

Engineering

First Advisor

Mahsa Ranji

Committee Members

Nader Sheibani, Adel Nasiri, Yongjin Sung, Yi Hu

Keywords

Disease Diagnosis, Fluorescence Microscopy, Image Analysis, Optical Coherence Tomography, Optical Cryoimaging, Oxidative Stress

Abstract

Optical imaging techniques have emerged as essential tools for reliable assessment of organ structure, biochemistry, and metabolic function. The recognition of metabolic markers for disease diagnosis has rekindled significant interest in the development of optical methods to measure the metabolism of the organ.

The objective of my research was to employ optical imaging tools and to implement signal and image processing techniques capable of quantifying cellular metabolism for the diagnosis of diseases in human organs such as eyes and lungs. To accomplish this goal, three different tools, cryoimager, fluorescent microscope, and optical coherence tomography system were utilized to study the physiological metabolic markers and early structural changes due to injury in vitro, ex vivo, and at cryogenic temperatures.

Cryogenic studies of eye injuries in animal models were performed using a fluorescence cryoimager to monitor two endogenous mitochondrial fluorophores, NADH (nicotinamide adenine dinucleotide) and FAD (flavin adenine dinucleotide). The mitochondrial redox ratio (NADH/ FAD), which is correlated with oxidative stress level, is an optical biomarker. The spatial distribution of mitochondrial redox ratio in injured eyes with different durations of the disease was delineated. This spatiotemporal information was helpful to investigate the heterogeneity of the ocular oxidative stress in the eyes during diseases and its association with retinopathy. To study the metabolism of the eye tissue, the retinal layer was targeted, which required high resolution imaging of the eye as well as developing a segmentation algorithm to quantitatively monitor and measure the metabolic redox state of the retina. To achieve a high signal to noise ratio in fluorescence image acquisition, the imaging was performed at cryogenic temperatures, which increased the quantum yield of the intrinsic fluorophores.

Microscopy studies of cells were accomplished by using an inverted fluorescence microscope. Fixed slides of the retina tissue as well as exogenous fluorophores in live lung cells were imaged using fluorescent and time-lapse microscopy. Image processing techniques were developed to quantify subtle changes in the morphological parameters of the retinal vasculature network for the early detection of the injury. This implemented image cytometry tool was capable of segmenting vascular cells, and calculating vasculature features including: area, caliber, branch points, fractal dimension, and acellular capillaries, and classifying the healthy and injured retinas. Using time-lapse microscopy, the dynamics of cellular ROS (Reactive Oxygen Species) concentration was quantified and modeled in ROS-mediated lung injuries. A new methodology and an experimental protocol were designed to quantify changes of oxidative stress in different stress conditions and to localize the site of ROS in an uncoupled state of pulmonary artery endothelial cells (PAECs).

Ex vivo studies of lung were conducted using a spectral-domain optical coherence tomography (SD-OCT) system and 3D scanned images of the lung were acquired. An image segmentation algorithm was developed to study the dynamics of structural changes in the lung alveoli in real time. Quantifying the structural dynamics provided information to diagnose pulmonary diseases and to evaluate the severity of the lung injury. The implemented software was able to quantify and present the changes in alveoli compliance in lung injury models, including edema.

In conclusion, optical instrumentation, combined with signal and image processing techniques, provides quantitative physiological and structural information reflecting disease progression due to oxidative stress. This tool provides a unique capability to identify early points of intervention, which play a vital role in the early detection of eye and lung injuries. The future goal of this research is to translate optical imaging to clinical settings, and to transfer the instruments developed for animal models to the bedside for patient diagnosis.

Available for download on Wednesday, January 03, 2018

Share

COinS