ORCID Identifier(s)

0000-0001-8649-3333

Graduation Semester and Year

2016

Language

English

Document Type

Dissertation

Degree Name

Doctor of Philosophy in Biomedical Engineering

Department

Bioengineering

First Advisor

Liping Tang

Second Advisor

Weihua Mao

Abstract

Supervising Professor: Liping Tang and Weihua Mao This study focuses on the development and analysis of novel computer vision algorithms for biomedical applications. Chapter 2 focuses on developing a 2D Canny edge-based DIR algorithm to register in vivo white light images taken at various time points. Using a mouse model, the accuracy of the canny edge-based DIR algorithm was tested using evaluation metrics, a ground truth synthetic scenario, and fluorescent gamma analysis. The results indicate that the Canny edge-based DIR algorithm performs better than all other algorithms. Chapter 3 focuses on the development of a novel longitudinal fluorescent signal tracking algorithm using an in vivo mouse model. To test the accuracy of the fluorescent signal tracking algorithm, confined mutual information was used. The results indicate that the automated fluorescent temporal implantation tracking algorithm performs better then rigid registration and DIR algorithms on tracking fluorescent signals in vivo. Chapter 4 aims to develop a 3D serial sectioning method to track the dispersion of model drugs via slow release devices in various ocular implantation sites. To achieve this, an automatic retinal delineation algorithm was developed and tested. The automatic sectioning algorithm was shown to be significantly indistinguishable from manual retinal sections. 3D serial sectioning retinal drug dispersion can be accurately modeled automatically. Chapter 5 is devoted to developing an automated landmark-guided deformable image registration (LDIR) algorithm between the planning CT and daily cone-beam CT (CBCT) with low image quality. The accuracy of the LDIR algorithm has been evaluated on a synthetic case and data of six head and neck cancer patients. The results indicate that LDIR performed better than all other algorithms. In general, all of the above computer vision solutions help improve upon, mimic, or automate the human visual system for biomedical applications.

Keywords

Deformable image registration, Computer vision, In vivo imaging

Disciplines

Biomedical Engineering and Bioengineering | Engineering

Comments

Degree granted by The University of Texas at Arlington

27139-2.zip (15371 kB)

Share

COinS
 
 

To view the content in your browser, please download Adobe Reader or, alternately,
you may Download the file to your hard drive.

NOTE: The latest versions of Adobe Reader do not support viewing PDF files within Firefox on Mac OS and if you are using a modern (Intel) Mac, there is no official plugin for viewing PDF files within the browser window.