Date of Award

12-2022

Document Type

Dissertation

Degree Name

Ph.D.

Department

Systems Engineering

Committee Chair

Silas, J, Leavesley, Ph.D.

Abstract

Hyperspectral imaging (HSI) has become a leading tool in the medical field due to its capabilities for providing assessments of tissue pathology and separation of fluorescence signals. Acquisition speeds have been slow due to the need to acquire signal in many spectral bands and the light losses associated with technologies of spectral filtering. Traditional methods resulted in limited signal strength which placed limitations on time sensitive and photosensitive assays. For example, the distribution of cyclic adenosine monophosphate (cAMP) is largely undetermined because current microscope technologies lack the combination of speed, resolution, and spectral ability to accurately measure Forster resonance energy transfer (FRET). The work presented in this dissertation assesses the feasibility of integrating excitation-scanning hyperspectral imaging methods in widefield and confocal microscopy as a potential solution to improving acquisition speeds without compromising sensitivity and specificity. Our laboratory has previously proposed excitation-scanning approaches to improve signal-to-noise ratio (SNR) and showed that by using excitation-scanning, most-to-all emitted light at each excitation wavelength band can be detected which in turn, increases the SNR. This dissertation describes development and early feasibility studies for two novel prototype concepts as an alternative excitation-scanning HSI technology that may xvi increase acquisition speeds without compromising sensitivity or specificity. To achieve this, two new technologies for excitation-scanning HSI were conceptually designed: - LED-based spectral illumination for widefield microscopy - Supercontinuum-laser-based spectral illumination for spinning disk confocal microscopy. Next, design concepts were theoretically evaluated and optimized, leading to prototype testing. To evaluate the performance of each concept, prototype systems were integrated with other systems and subsystems, calibrated and feasibility assays were executed. This dissertation is divided into three main sections: 1) early development feasibility results of an excitation-scanning widefield system of systems prototype utilizing LED-based HSI, 2) Excitation-scanning HSI and image analysis methods used for endmember identification in fluorescence microscopy studies, and 3) early development feasibility of an excitation-scanning confocal SoS prototype utilizing a supercontinuum laser light source. Integration and testing results proved initial feasibility of both LED-based and broadband-based SoSs. The LED-based light source was successfully tested on a widefield microscope, while the broadband light source system was successfully tested on a confocal microscope. Feasibility for the LED-based system showed that further optical transmission optimization is needed to achieve high acquisition rates without compromising sensitivity or specificity. Early feasibility study results for the broadband-based system showed a successful proof of concept. Findings presented in this dissertation are expected to impact the fields of cellular physiology, medical sciences, and clinical diagnostics by providing the ability for high speed, high sensitivity microscopic imaging with spectroscopic discrimination.

Share

COinS