Document Details

Document Type : Thesis 
Document Title :
EARLY SKIN MELANOMA DETECTION AND CLASSIFICATION USING CAD SYSTEM
الكشف المبكر عن سرطان الجلد الميلانيني وتصنيفه باستخدام نظام التشخيص بمساعدة الحاسوب
 
Subject : Faculty of Engineering 
Document Language : Arabic 
Abstract : Skin cancer regarded as a highly risky tumor, which causes death worldwide. The main reason that causes skin cancer is the overexposure sunlight Ultraviolet Radiation (UVR). UVR causes damage to DNA in the skin. According to American cancer society (ACS) of incidence rates of melanoma predicted as 96000 between males and females in the United States for 2019. The skin cancer originated from the pigmented melanin lesion in the skin. The dermatologist may face difficulties in classifying the skin lesions. The early melanoma prognosis may reduce the ability of cancer to spreading to other organs by the blood that causes deaths. We suggested in our study to investigate the feature extraction using wavelet packet transform (WPT) in combination with an entropy method. A probabilistic neural network (PNN) was suggested for classification between different classes of lesions. Several experiments were conducted to determine the best results of recognition rates (Accuracy). Our methodology achieved a relatively high recognition rate of around 86%. 
Supervisor : Prof. Khaled Daqrouq 
Thesis Type : Master Thesis 
Publishing Year : 1442 AH
2020 AD 
Added Date : Friday, December 25, 2020 

Researchers

Researcher Name (Arabic)Researcher Name (English)Researcher TypeDr GradeEmail
معن صبحي المرعيAl-marei, Maen SubhiResearcherMaster 

Files

File NameTypeDescription
 46824.pdf pdf 

Back To Researches Page