Electronic Theses and Dissertations

Identifier

277

Author

Sreya Ghosh

Date

2011

Date of Award

4-19-2011

Document Type

Thesis

Degree Name

Master of Science

Major

Electrical and Computer Engr

Concentration

Electrical Engineering

Committee Chair

Chrysanthe Preza

Committee Member

Eddie Jacobs

Committee Member

Aaron Robinson

Abstract

Noninvasive three-dimensional (3D) imaging of thick samples in fluorescence microscopy is traditionally achieved via the method of optical sectioning where multiple two-dimensional images of an object acquired while focusing at different depths within the object. Defocus, spherical aberrations (SA) and photo bleaching are known to affect the accuracy achieved in 3D images. Double-Helix point spread functions (DH-PSF), the result of PSF engineering, have been used for super localization of point sources in 3D samples. The unique DH-PSF design features 2 dominant lobes in the image plane which appear to rotate as the axial (z) position of a point light source with respect to the imaging lens is changed. Thus, the angular orientation of the DH-PSF lobes encodes the imaging depth. The DH-PSF lobes are horizontal when the emitter is in focus. As the emitter is moved toward the objective lens, the DH-PSF lobes rotate, however, if the emitter is moved away from the objective the lobes rotate in the opposite direction. In this thesis, the effect of the DH-PSF in 3D optical sectioning microscopy is studied for different imaging conditions using simulations that introduce SA and noise. The frequency domain features of the DH-PSF were studied and were shown to preserve the rotation characteristic that encodes depth information. In addition frequency analysis shows that the DH-PSFs are less sensitive to defocus and SA at small imaging depths than the conventional PSF. The effect of SA is shown to cause an error in the super localization of emitters which it was successfully corrected by a new methodology developed in this thesis. Different approaches to estimate intensity and location of structures from computer generated images were investigated. The objects were effectively recovered from the images with and without SA and noise. Thus from this study we conclude that DH-PSF based systems can be effectively integrated into the conventional fluorescent microscopy system to help in particle tracking, super localization and object ranging.

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.

Share

COinS