Volumetric Breast Density Analysis for Digital Mammograms
| Autor: | Liang-Chun Yin, 尹亮鈞 |
|---|---|
| Rok vydání: | 2010 |
| Druh dokumentu: | 學位論文 ; thesis |
| Popis: | 98 Breast cancer is one of the most common malignancies in women and it continues to be a major cause of death among women worldwide. Several studies have shown that the women with dense breasts are at higher risk than the women whose breasts are less dense. Hence, breast density has to be accepted as a clinically highly significant predictor of breast cancer risk. Most studies of breast density are performed on the mammographic images. Because the two-dimensional (2-D) mammographic density shows large variability, the three-dimension (3-D) breast magnetic resonance imaging (MRI) has been used in some recent studies for achieving more reasonable consistency. In this paper, we propose a three-dimension (3-D) method to estimate the breast density with digital mammograms. This method is based on the assumption that the composition of the breast contains only two types, the fat tissue and the fibroglandular tissue. The linear attenuation coefficients of these tissues are used to obtain the composition at a given pixel. The density estimations of the mammographic images are compared with the results in the MRI. The validation between the MRI and the digital mammograms using 17 cases has been performed and shows a high correlation. Also, the proposed volumetric breast density estimation using the mammographic images is applied to evaluate the breast density change for 39 patients with the chemoprevention treatments. In 32 of 39 cases, the fibroglandular volume after treatment is reduced and the fibroglandular volume reduction is 13.40%. However, there are 7 cases with their fibroglandular volume increasing of 5.72% after the chemoprevention and 6 of 7 increasing cases receive shorter treatment (less than 12 months). |
| Databáze: | Networked Digital Library of Theses & Dissertations |
| Externí odkaz: |
|