Interpretation of Mammographic Using Fuzzy Logic for Early Diagnosis of Breast Cancer
Jorge Raúl Pérez Gallardo, Beatriz Hernández‐Vera, Alberto Alfonso Aguilar Lasserre, Rubén Posada‐Gómez · 2008
Accuracy and interpretability are two important objectives in the design of Fuzzy Logic model. In many real-world applications, expert experiences usually have good interpretability, but their accuracy is not always the best. Applying expert experiences to Fuzzy Logic model can improve accuracy and preserve interpretability. In this study we propose an accessible tool that helps medical interpretation of suspect zones or tumors in mammographics. This paper describes a methodology to locate precisely different kind of lesions in breast cancer patients. The use of Fuzzy Logic model improves the diagnostic efficiency in tumor progression. After applying an image segmentation method to extract regions of interest (ROIs), the values obtained feed the system. The Fuzzy Logic model processes them to achieve Breast Imagine Reporting And Data System (BI-RADS®). Some experimental results on breast images show the feasibility of the propose methodology