Masters Theses


Jixiang Chen


"This thesis focuses on the development and implementation of color analysis of melanoma images to make a diagnosis.

Melanoma is a form of skin cancer, which can be deadly if not treated in the early stages. Current melanoma detection techniques rely on guidelines, like asymmetry, border irregularity, color and diameter of the tumor, called the ABCD criteria [1]. Among them, color analysis becomes an important index in the diagnosis of malignant melanoma. It is also hypothesized that the clustering of melanoma-colored pixels occurs more intensely in melanomas than in benign lesions.

It is believed that melanoma colors and color clustering may have some particular characteristics in the boundary region of malignant melanoma. The main objectives of this research are to determine if there are some special features in the boundary region of the melanoma, and to try to improve the diagnostic accuracy based on its features. By processing 50 clinical melanoma images and 50 clinical benign images, the research finally indicates that the diagnostic result is more encouraging if we analyze the image based on melanoma colors or clustering in the offset boundary that only encompasses a certain percentage of the total tumor size"--Abstract, page iii.


Moss, Randy Hays, 1953-

Committee Member(s)

Stanley, R. Joe
Fu, Yongjian


Electrical and Computer Engineering

Degree Name

M.S. in Computer Engineering


University of Missouri--Rolla

Publication Date

Fall 2001


vii, 50 pages

Note about bibliography

Includes bibliographical references (page 49).


© 2001 Jixiang Chen, All rights reserved.

Document Type

Thesis - Restricted Access

File Type




Subject Headings

Melanoma -- Diagnosis

Thesis Number

T 8001

Print OCLC #


Link to Catalog Record

Electronic access to the full-text of this document is restricted to Missouri S&T users. Otherwise, request this publication directly from Missouri S&T Library or contact your local library.

Share My Thesis If you are the author of this work and would like to grant permission to make it openly accessible to all, please click the button above.