1Research Scholar, Department of Computer Science and Engineering, Sathyabama Institute of Science and Technology, Chennai, Tamilnadu, India
2Scientist-D & Assistant Professor, Department of Computer Science and Engineering, Sathyabama Institute of Science and Technology, Chennai, Tamilnadu, India
Online published on 21 July, 2018.
One of the most dangerous forms of skin cancer is melanoma which increases the mortality rate compared to other form of skin cancers. It occurs in the skin typically, but occurs in eye, mouth rarely. The main treatment for skin cancer is surgery and the survival rate is very low if spread occurs. Mortality rates can be reduced by early detection through screening. The abnormality in skin can be diagnosed by a dermatology specialist through the interpretation of the Dermoscopic images. However, the rate of correct diagnosis by experts is estimated around only 75–84%. In this paper, recent advances for skin cancer diagnosis using computer aided diagnosis by features analysis and machine learning are reviewed. A comparative analysis of recent approaches in terms of sensitivity and specificity is also made.
Skin cancer, Melanoma, Computer aided diagnosis, Machine learning, Feature analysis
