A Deep CNN Model for Skin Cancer Detection and Classification

Abstract

Skin cancer is one of the most dangerous forms of cancer when it is not detected early. If timely diagnosis and treatment are not provided, it can spread to other parts of the body, making it more difficult to treat. If early detection happens at that time, it plays a critical role in saving a life, as such an automated system for Skin lesion recognition has a highly valuable role in saving time and efforts, but also reducing the burden of the professional. This paper focus on using CNNs to classification skin cancer of 8 type and also provide difference between cancer skin and normal skin utilizing CNNs system processes the skin images to identify the various type and conditions including the a keratosis, bcc, dermatofibroma, meloma, nevus, pgk,  scc, vascular lesion the deep learning model demonstrate a classification accuracy of 79.80%, model design with the multiple convolution layer, pooling layer, batch normalization, Adam optimization, max pooling layer, and output classification used SoftMax layer.