SKIN CANCER DETECTION USING DEEP LEARNING

Abstract

In this work, an intelligent system for skin cancer detection is proposed. It is known that skin cancer is one of the deadliest diseases in the world, so it must be detected correctly in the early stages itself for saving human life, for which a fully automatic system using deep learning techniques can be used.In this work, HAM10000 dataset is used for classification which contains classes of skin cancer images. Deep Convolutional Neural Network, AlexNet, VGG-16 and Inception V3 are tested for classification of skin lesions. A dedicated model is proposed for skin cancer classification, Skin Cancer Detection Model, and the model is tested. The same dataset is trained with all the above models and also evaluated the known quantitative measures such as accuracy, precision and recall. Finally, the confusion matrix of the models is plotted. The skin cancer detection model achieved better accuracy than other models. Skin detection model has an accuracy, precision, recall and F1 score of 97.354%, 98%, 97%, 97% respectively