A CNN Approach for Skin Cancer Classification

Abstract

Skin cancer is one of the most common types of cancer in the world. In the US, there are around 5 million new people diagnosed by skin cancer each year. Skin cancer is divided into five stages based on the spreading of the disease with survival rates between 99% at the first stage and 20% at the fifth stage. Therefore, early diagnosis of skin cancer is crucial in saving human lives. The shape and the texture of the skin for different skin cancer lesions are different. Therefore, image processing techniques would be beneficial in providing diagnostic aid for the physicians when diagnosing skin cancer. In this paper we will explore the usage of image classification algorithms to identify skin cancer types by using Convolution Neural Networks (CNN). The proposed algorithm is applied on a dataset that consists of 10,000 images of seven different types of lesions.