A Hybrid System for Classification of Skin Cancer Images Using Artificial Neural Network and Support Vector Machine

Abstract

Skin cancer is a very serious disease and identifying it early is therefore very important to take the appropriate treatment before the situation worsens. Melanoma cancer is one of the most prevalent malignant types in recent years because it spreads to the rest of the body quickly, it must be detected early and treated. Computer Vision in medical imaging analysis is important in the diagnosis of various diseases at many stages. These methods give a high-resolution assessment of the disease by analyzing the images digitally. The steps followed in this paper involve collecting an image dataset from a global website for skin cancer about 1500 photos of different sizes, preprocessing, segmentation using threshold, feature extraction for Asymmetry, Border, Color, Diameter, (ABCD), etc., calculating the total of dermatoscopy score and then classification using Artificial Neural Network (ANN) and Support Vector Machine (SVM) in the same of the stage. The results of the system display that the attained classification accuracy is 98.38%