Analysing the Performance of Classifiers for the Detection of Skin Cancer with Dermoscopic Images

Skin cancer is one of the major causes of deaths in recent days. Early detection of skin cancer reduces death at higher rate. Ceroscopy is one of the major modalities used in diagnosis of skin lesions. Skin lesions are of different types. Among them the most common types of skin lesion found in human are melanoma, basal cell carcinoma (BCC) and squamous cell carcinoma (SCC).The accurate diagnosis information cannot be obtained by human interpretation. In order to overcome the error due to human interpretation an efficient computerized image analysis system has been developed. The proposed image analysis system consists of preprocessing, lesion segmentation, feature extraction and classification. In classification, different types of classifiers such as support vector machine (SVM), probabilistic neural network (PNN) and adaptive neurofuzzy inference system (ANFIS) are applied to classify the skin cancer types and their performance is compared using the evaluated parameters. Citation: Kavimathi.P, Sri Venkateswara College Of Engineering; Sivagnanasubramanian.S.P ,. "Analysing the Performance of Classifiers for the Detection of Skin Cancer with Dermoscopic Images." Global Research and Development Journal For Engineering : 437 - 442.