Information extraction algorithms based on optical coherence tomography data

In this dissertation, we present several algorithms developed for optical coherence tomography (OCT) datasets. We show that by incorporating physical modeling, mathematical analysis and signal processing, it is possible to uncover the huge potential hidden in the raw OCT data. The automated Interferometric Synthetic Aperture Microscopy algorithm uses image sharpness metrics to automatically search for the optimal parameters for each step of the processing, resulting in near-optimal reconstruction comparable to images manually tuned by experts. The filtering algorithm for Doppler OCT images combines mathematical optimization with GPU parallel programming, and drastically increases the dynamic range of Doppler OCT velocity maps, making Doppler OCT potentially applicable to faster blood vessels such as the carotids. The susceptibility tensor ISAM algorithm converts the complicated multipolarization scattering and detection problem into a linear algebra problem and solves it through regularized least square method, which enables us to indirectly measure all nine components of the susceptibility tensor for clusters of discrete particles. We also discuss the potential method to extend susceptibility tensor ISAM to continuous samples, which promises better contrast between tissue types and insights into the anisotropic structure inside biological tissue.

Name thesis:degree_name

Ph.D.

Level thesis:degree_level

Dissertation

Discipline thesis:degree_discipline

Electrical & Computer Engr

Grantor

University of Illinois at Urbana-Champaign

Year dc:date

2018

Author dc:creator

• Xu, Yang

Contributors dc:contributor

• Carney, P. Scott
• Eden, J. Gary
• Singer, Andrew C.
• Oelze, Michael L.
• Gruev, Viktor

• Optical coherence tomography
• Interferometric synthetic aperture microscopy

Statement dc:rights

• Copyright 2017 Yang Xu

Language dc:language

en

Handle dc:identifier

http://hdl.handle.net/2142/99193

OAI identifier oai:identifier

oai:www.ideals.illinois.edu:2142/99193