Electronic Theses and Dissertations

Identifier

6074

Date

2017-12-07

Document Type

Dissertation (Campus Access Only)

Degree Name

Doctor of Philosophy

Major

Electrical and Computer Engr

Concentration

Electrical Engineering

Committee Chair

Chrysanthe Preza

Committee Member

Eddie Jacobs

Committee Member

Omar Skalli

Committee Member

Ana Doblas

Abstract

Three-dimensional (3D) fluorescence microscopy (FM) is an integral part of biomedical research as it provides the necessary tool to investigate the interaction between the biomolecules in multiple dimensions (i.e. spatial, spectral, and temporal). This dissertation contributes toward two special types of FM: (i) computational optical sectioning microscopy (COSM), in which computational methods are used to improve the performance of the conventional imaging system, and (ii) structured illumination microscopy (SIM), which is used to increase the resolution of the optical microscope beyond the diffraction limit. A simplifying assumption of COSM is that the point spread function (PSF) does not change throughout the imaging volume; however, this assumption is not valid for optically thick samples (>5 µm), if the refractive index (RI) between the sample-mounting medium and the microscope objective (MO) lens’ immersion medium is different. Such cases require the use of computationally intensive depth-variant (DV) image restoration methods to avoid artifacts. To overcome this challenge, a wavefront encoded imaging system is developed, where we have demonstrated through simulation and experiment that a PSF does not change significantly up to a 60-µm imaging depth, which consequently improves the computational efficiency while restoring optically thick samples. Another contribution of this dissertation is investigating the impact of SA on SIM and the use of wavefront encoding to reduce the effect of SA on the SIM system. Data from the proof-of-concept setup was acquired and compared to the simulation to validate the implementations. Preliminary results demonstrate challenges that need to overcome, in order to be able to assess the impact of this approach on addressing SA satisfactorily. As an alternative approach, an image restoration method is developed and evaluated to improve the performance of SIM when the fringe visibility of the structured light is low, a condition that occurs when the sample is optically thick and/or the modulation frequency is high. Restoration results from simulated and experimental SIM raw images show improved signal to noise ratio (SNR) and adequate optical sectioning in 3D images, in which more details of fine structures are evident compared to results obtained with two existing computational methods.

Comments

Data is provided by the student.

Library Comment

Dissertation or thesis originally submitted to the local University of Memphis Electronic Theses & dissertation (ETD) Repository.

Download

Share

COinS