Convolutional Neural Networks for Classifying Melanoma Images

Abstract

In this work, we address the problem of skin cancer classification using convolutional neural networks. A lot of cancer cases early on are misdiagnosed leading to severe consequences including the death of patient. Also there are cases in which patients have other problems and doctors interpret it as skin cancer. This leads to unnecessary time and money spent for further diagnosis. In this work, we address both of the above problems using deep neural networks and transfer learning architecture. We have used publicly available ISIC databases for both training and testing our network. Our model achieves an accuracy of 0.935, precision 0.94, recall 0.77, F1 score 0.85 and ROC-AUC 0.861 which is better than the previous state of the art approaches.