Masters Theses

Keywords and Phrases

Benign; Dermis; Image features; Image processing; Melanoma; Skin cancer


""Detection of melanoma remains an empirical clinical science. New tools for automatic discrimination of melanoma from benign lesions in digitized dermoscopy images may allow an improvement in early detection of melanoma. This research implements a fast version of the median split algorithm in an open source format and applied to four-color splitting of the lesion area to capture the architectural disorder apparent in melanoma colors. This version of the median split algorithm splits colors along the color axis with maximum range". For a dermoscopy set of 888 images, K-means clustering algorithm is compared with a median split algorithm to find which model is performing better according to logistic regression analysis from SAS. For images with the median split algorithm, a full model of 208 features and a robust model of 45 features were developed for an 837 dermoscopy image set and a threshold was selected using logistic regression analysis that shows the most important features in both the models. Using this threshold, we checked the robustness and accuracy on a test model of 78 dermoscopy images with full and robust model. The median split algorithm is fast, requiring less than one second per image and only a four-color splitting, but it captures sufficient critical information regarding color disorder, with peripheral inter-color boundaries showing the highest significance for melanoma discrimination"--Abstract, page iii.


Moss, Randy Hays, 1953-

Committee Member(s)

Shrestha, Bijaya
Stanley, R. Joe


Electrical and Computer Engineering

Degree Name

M.S. in Electrical Engineering


Missouri University of Science and Technology

Publication Date



xi, 179 pages

Note about bibliography

Includes bibliographical references (pages 177-178).


© 2013 Venkata Sai Narasimha Kaushik Ghantasala, All rights reserved.

Document Type

Thesis - Open Access

File Type




Thesis Number

T 10995

Electronic OCLC #