Automatic Diagnosis of Skin Cancer Using Neural Networks

The paper presents an automated classification system for melanoma diagnosis. It is based on a convolutional neural network that is fed with images of skin lesions which are preprocessed in advance. The preprocess step is necessary for reducing the number of artifacts present in the images and hence, maximize the classification accuracy. The proposed solution is based on the training of the neural network with a series of preprocessed clinical images, classifying them in two categories: benign or malignant.