SKIN LESION CLASSIFICATION FROM DERMOSCOPY AND CLINICAL IMAGES WITH A DEEP LEARNING APPROACH

Computer-aided diagnosis (CAD) systems based on deep learning approaches are now feasible due to the availability of big data and powerful computational resources.The medical image-based CAD great interest in numerous diseases, but especially for skin cancer diagnosis, models have been mostly developed dermoscopy images. Models clinical images few, mainly unavailability volumes relevant data. However, able classify lesions from would be valueboth population clinicians as an initial early screening that leadpatients visiting a dermatologist case suspicious lesions. This is even more pronounced areas where there lack instruments. Thus, this paper, we aimed build classifier bothdermoscopy discriminate The classification made among three benign two malignant categories, which include Nevus, Benign not nevus, malignancy, Melanoma Non-Melanocytic Carcinoma.The proposed achieves Area Under Curve ranging between 0.75 0.9 five examined categories.