A New Fat-Removal-Based Preprocessing Pipeline for MLO View in Digital Mammograms

Specific anatomical structures from the female body, such as axillary slope, armpit, pectoral muscle, or abdominal tissue, can be present in mammograms and might affect proper mammogram analysis, especially populations with overweight issues, is case Mexico. This work aims to determine if better results obtained an automatic analysis by removing fatty tissues a preprocessing step. The experimentation carried out applying muscle segmentation technique Mexican dataset comparing same mammograms. Furthermore, performed using Portuguese public datasets, INbreast BCDR. conducted experiment will allow us differences across different populations. fat removal method based on breast contour-edge which points of interest are detected cut tissue. Our suggest that proposed suitable technique, obtaining 94.18% acceptance, according qualitative performed, showed tissue yields contains significant patients.