Abstract: Mammography is the best existing examination for the detection of early signs of breast cancer such as masses, calcifications, bilateral asymmetry and architectural distortion. In addition, mammograms are difficult to interpret, especially of cancer at their early stages. However, the presence of artifacts and noises can disturb the detection of breast cancer and reduce the rate of accuracy in the computer aided analysis (CAD). For this cause, pre-processing of mammogram images is very important in the process of breast cancer analysis because it could reduce the number of false positive. On the other hand, it could help radiologists to make a comparison between mammograms. The objective of this paper is to propose an algorithm of pre-processing on medio-lateral oblique-view (MLO) mammograms that involves two stages: the first is image enhancement of which the aim is to improve the interpretability or perception of information in images for human viewers, or to provide a better input for other automated image processing techniques. Eight techniques for the enhancement of overall mammograms inside the mini-MIAS database were implemented. The results were evaluated using Peak Signal to Noise Ratio (PSNR). The second stage is to extract the breast region from the rest of the image (background). For this purpose, a method based on automatic thresholding was used. Furthermore, the proposed pre-processing method was evaluated on 322 breast images collected from MIAS mini-database. The enhancement and extraction of the breast profile proved promising findings.
Keywords: Breast region extraction, Segmentation, Mammogram thresholding, Contrast stretching enhancement, Median filter and Computer aided design.
Title: Preprocessing Technique for Mammographic Images
Author: Luqman Mahmood Mina, Nor Ashidi Mat Isa
International Journal of Computer Science and Information Technology Research
ISSN 2348-120X (online), ISSN 2348-1196 (print)
Research Publish Journals
