Masters Theses

Author

Cheng Lu

Abstract

"Cervical cancer, the second most common cancer affecting women worldwide and the most common in developing countries can be cured in almost all patients, if detected early and treated. However, cervical cancer incidence and mortality remain high in resource-poor regions, where early detection systems often cannot be maintained because of inherent complexity. The National Cancer Institute (NCI) has collected a vast amount of visual information, 100,000 cervigrams (35 mm color slides), screening thousands of women by this technique. In addition to the cervigrams, large digitized histology images are being archived.

In this research, a framework for automatic recognition and classification of cervical intraepithelial neoplasia has been developed. Data sets of 62 image sets with segmented squamous epithelium regions were obtained from the National Library of Medicine, which were analyzed using the framework developed.

This thesis presents methods used in this research to improve the classification results by implementing different feature extraction algorithm and classification algorithm. A leave-one-image-out approach was explored and yielded an overall classification rate as high as 72.58% for exact classification scoring using the cervical intraepithelial neoplasia (CIN) classes Normal, CIN1, CIN2, and CIN3."--Abstract, page iii.

Advisor(s)

R. Joe Stanley

Committee Member(s)

Moss, Randy Hays, 1953-
Stoecker, William V.

Department(s)

Electrical and Computer Engineering

Degree Name

M.S. in Computer Engineering

Sponsor(s)

National Library of Medicine (U.S.)

Publisher

Missouri University of Science and Technology

Publication Date

2013

Pagination

vii, 39

Note about bibliography

Includes bibliographical references (page 38).

Rights

© 2013 Cheng Lu, All rights reserved.

Document Type

Thesis - Open Access

File Type

text

Language

English

Library of Congress Subject Headings

Cancer -- Histopathology
Cervix uteri -- Histopathology
Pattern recognition systems -- Classification
Image analysis
Cervix uteri -- Cancer -- Diagnosis

Thesis Number

T 10998

Electronic OCLC #

1002219154

Share

 
COinS