Influência da aplicação de filtros digitais no contraste de imagens mamográficas e de tomossíntese

Breast cancer is a major public health problem with high occurrence worldwide. Mammography is the main radiological exam in the screening of this cancer, and recently, the emergence of tomosynthesis as an additional exam has been helping in the diagnosis of the disease. However, some images do not have a good quality, mainly due to the high presence of noise in, requiring the application of digital processing methods. Thus, the goal of this work is to compare the effects of Wiener and Mean filters on phantom images acquired in different equipment from the contrast-noise ratio (CNR) calculations. The results indicated an increase of CNR values for all original images when applying digital filters, increasing the image quality. Moreover, comparing the original images, it was noted that the highest CNR value was obtained in the tomosynthesis image, proving a greater contrast due to the acquisition geometry of the images of this feature. Thus, we concluded that the Mean filter was the one that obtained the highest CNR value, however, this type of filter causes a blur in the image when compared to the Wiener filter, indicating the importance of a visual analysis in conjunction with the quantitative one.