The project is based on pattern analysis for early diagnosis of skin cancer. To this end, digital image processing techniques will be developed to help diagnose pigmented lesions of the skin using pattern analysis of dermatoscopic images.
More specifically, the project consists of the development of a tool that uses algorithms capable of recognising diagnostic patterns of pigmented lesions appearing on the skin, and so automatically screens patients with a high risk of developing skin cancer and differentiates them from those that do not. This will reduce the number of patients referred from primary to specialist care and enhance the effectiveness of treatment applied to this kind of melanoma by identifying the disease at a very early stage.
A database of real, multispectral dermatoscopic images previously screened by specialists will be used for training doctors to use the system