Analyzing Digital Image by Deep Learning for Melanoma Diagnosis

摘要: Image classification is an important task in many medical applications, order to achieve adequate diagnostic of different lesions. Melanoma a frequent kind skin cancer, which most them can be detected by visual exploration. Heterogeneity and database size are the difficulties overcome obtain good performance. In this work, deep learning based method for accurate wound regions proposed. Raw images fed into Convolutional Neural Network (CNN) producing probability being melanoma or non-melanoma. Alexnet GoogLeNet were used due their well-known effectiveness. Moreover, data augmentation was increase number input images. Experiments show that compared models high performance terms mean accuracy with very few without any preprocessing.