Skin Cancer Detection and Diagnosis Using Image Processing and Implementation Using Neural Networks and ABCD Parameters

Skin cancer is most dangerous form of all cancers that occurs for human being. Among these skin cancers melanoma is most deadliest form which accounts more than 40% of all over world, Even though melanoma is deadliest if detected in early stages then it can be treated successfully. Day-by-Day many computer based automatic detection and diagnosis of skin cancer are developing. In this paper we present an improved method using image processing method which uses statistical features and dermoscopic features such as ABCD(Asymmetry, Border, Color and Diameter) for detection and diagnosis. The steps involved in this method are creation of database for dermoscopic images, filtering for noise removal, segmentation using thresholding, statistical feature extraction using Gray Level Co-occurrence Matrix (GLCM), Calculating Total Dermoscopy Score and then classification using Artificial Neural Network (ANN). The combined results of the neural network and the ABCD parameters indicate that the developed methodology proved effective and efficient for the skin cancer detection and diagnosis.