Sistema embarcado para suporte ao diagnóstico de microcalcificações em mamografias digitais

International studies indicate that breast cancer is the second greatest cause of death in women with cancer worldwide, and the best way to treat this disease is by its early diagnosis, which can be performed by analyzing images of digital mammograms. One of the ways to make the diagnosis using images is to look for microcalcifications, small accumulations of calcium that can be classified into five types, according to the Le Gal classification scheme. This work presents an embedded system developed to classify microcalcifications according to the Le Gal classification scheme using wavelets and artificial neural networks. The system was developed using a Nios II processor by the configuration of the designed hardware on the FPGA located on the DE2 -115 board. Tests with simulated images showed an accuracy rating of 92.50%. Some tests with real images also showed good results in classification of microcalcifications. All of this indicates the possibility of using embedded devices for the early diagnosis of breast cancer.