DocumentCode
2046846
Title
Bilateral Breast Volume Asymmetry in Screening Mammograms as a Potential Marker of Breast Cancer: Preliminary Experience
Author
Eltonsy, Nevine H. ; Elmaghraby, Adel S. ; Tourassi, Georgia D.
Author_Institution
Louisville Univ., Louisville
Volume
5
fYear
2007
fDate
Sept. 16 2007-Oct. 19 2007
Abstract
The biological concept of bilateral symmetry as a marker of developmental stability and good health is well established. Although most individuals deviate slightly from perfect symmetry, humans are essentially considered bilaterally symmetrical. Studies have shown that if an individual is exposed to genetic mutations or environmental stresses, the homeostatic mechanisms that maintain symmetry of paired structures (such as breasts) tend to break down. Consequently, increased fluctuating asymmetry of paired structures could be an indicator of poor health. This preliminary study tested if bilateral morphological breast asymmetry in screening mammograms correlates with the presence of breast cancer. Following the biological definition of breast asymmetry in terms of volume, we applied automated computer algorithms for screening mammograms that segment the breast region and then measure each segmented breast´s volume. These parameters were measured separately for each breast in each mammographic view (CC and MLO). Then, the normalized absolute differences of these parameters were investigated as measurements of fluctuating asymmetry (FA). Based on 268 cancer cases and 82 normal cases from the DDSM database, we observed that cancer patients demonstrate statistically significantly higher fluctuating asymmetry in their screening mammograms than patients with normal screening studies. Using an artificial neural network to combine FA measurements from both views along with the patient´s age and breast parenchymal density resulted in an ROC area of 0.80plusmn0.03. These results suggest that bilateral breast volume asymmetry estimated in screening mammograms should be studied as a risk factor for breast cancer.
Keywords
cancer; image segmentation; mammography; medical image processing; neural nets; artificial neural network; automated computer algorithm; bilateral breast volume asymmetry; breast cancer; homeostatic mechanism; image segmentation; mammograms; Biology computing; Breast cancer; Databases; Delta-sigma modulation; Genetic mutations; Humans; Stability; Stress; Testing; Volume measurement; mammography; quantitative image analysis; risk assessment;
fLanguage
English
Publisher
ieee
Conference_Titel
Image Processing, 2007. ICIP 2007. IEEE International Conference on
Conference_Location
San Antonio, TX
ISSN
1522-4880
Print_ISBN
978-1-4244-1437-6
Electronic_ISBN
1522-4880
Type
conf
DOI
10.1109/ICIP.2007.4379751
Filename
4379751
Link To Document