Basic feature extractions from mammograms

Brest cancer is the most frequent cause of cancer-induced deaths in women in Europe. Systematic early detection through screening, effective diagnostic pathways and optimal treatment have the ability to substantially lower current breast cancer mortality rate. To produce image of the internal breast structure (mammogram) with adequate quality, each part of the imaging chain must function properly. Nowadays, image processing algorithms play a significant role in enhancements and visibility of specific image details. In this work, digitizing mammograms were analyzed using basic point operators to highlight particular features and to extract quantitative information. Main focus was to the contrast enhancement between “suspicious” breast structures and adjacent tissues. Basic algorithms such as normalization, equalization and thresholding were applied and their actions have been shown in both: the image and its histogram. Matlab as an image processing tool was used. It was shown that basic point operators could be successfully used to help radiologists by increasing probability in early detection of breast cancer, even if the original image was not optimally taken.