Mammography image processing to evaluate image quality.

C. QUINTANA; S. OJEDA; G. TIRAO; M. VALENTE

Foz de Iguazu

Simposio; III SEEMI – Simpósio em Estatística Espacial e Modelagem de Imagens; 2010

Universidade Estadual do Oeste do Paraná, Brasil

Mammography imaging proved to be the best non invasive method for breast cancer diagnosis but it requires that irradiation parameters to be set within Protocols recommendations (minimal dose delivering). This work presents a whole investigation about mammography image formation by means of validated Monte Carlo simulations along with further image analysis and processing. Four different image processing methods have been suitably introduced and investigated according to their capability for micro-calcification detection. The obtained results suggest the feasibility of all the proposed method. Furthermore, it was possible to characterize the reliability of each one and to infer the corresponding advantages/disadvantages.