Masters Theses

Beibei Cheng

Abstract

"The blood vessels are part of the circulatory system and function to transport blood throughout the body. Vessels have their own features such as distinctive color compared to surrounding skin as well as distinctive curved and/or linear shape. Telangiectases are small dilated blood vessels near the surface of the skin or mucous membranes, measuring between 0.5 and 1 millimeter in diameter. In this research, image analysis techniques are investigated to detect vessels in dermoscopy skin lesion images. Machine vision and neural network methods are explored to discriminate skin lesions containing telangiectases from those containing normal vessels. A vessels Detection technique is implemented firstly to find the possible vessels in dermatology skin lesion images. In addition, a noise filtering technique is applied, which filters out the "noise" such as hair, bubble and so one, according to their own features. Based on the fact that some of the images are fuzzy, a contrast enhancement technique can be added to increase the contrast. After obtaining the final masked regions containing vessel-like structures, features are computed to facilitate the discrimination of skin lesion with normal vessels from lesions containing telangiectases. The features are mostly about the number, shape and size of telangiectases mask"--Abstract, page iii.

Advisor(s)

Stanley, R. Joe

Committee Member(s)

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

Department(s)

Electrical and Computer Engineering

Degree Name

M.S. in Computer Engineering

Sponsor(s)

Stoecker and Associates

Publisher

Missouri University of Science and Technology

Publication Date

Spring 2009

Pagination

ix, 47 pages

Rights

© 2009 Beibei Cheng, All rights reserved.

Document Type

Thesis - Open Access

File Type

text

Language

English

Subject Headings

Basal cell carcinoma -- Diagnosis -- Computer programsImage processing -- Computer programsSkin -- Cancer -- Diagnosis -- Computer programsTelangiectasia

Thesis Number

T 9480

Print OCLC #

436057808

Electronic OCLC #

318345805

Download

Share

 
COinS