Modelagem e Implementação de um Phantom Digital de Mama Baseado em Ruído de Perlin

Abstract

Breast cancer is a disease that affected over 2 million women worldwide in 2022, with approximately 94,000 cases diagnosed in Brazil. The number of cases and deaths due to this cancer may increase in the coming years due to population aging and growth. When diagnosed early, patients have a higher chance of being cured, which implies lower breast cancer mortality rates. The gold standard for breast cancer detection is the mammography exam, primarily recommended for individuals aged 50 to 69. Ordinance No. 2,898, from November 28, 2013, established the Programa Nacional de Qualidade em Mamografia (PNQM), aimed at implementing quality control of mammography across the national territory. Given the necessity of quality control in mammography exams, the objective of this study was to develop a realistic breast phantom simulator, which can be used for virtual clinical trials to support diagnostic tool testing, professional training, and research on breast models. A graphical interface was developed that allows users to create diverse digital breast phantoms based on parameters chosen by the user. The simulator models the breast, nipple, mammographic background, and glandular tissue to represent breast structures. Each generated image can be saved in DICOM format for later access, which can be used for image processing tests. This simulator provides visual results close to real images, though it has some limitations regarding the diversity of models produced, resulting in low variability in breast density. The results indicate a need to represent greater variation in breast density in future work, as well as additional testing to validate the quality of the simulator.