Breast cancer is the most prevalent non-skin cancer in women in the US. To increase the early detection of breast cancer, annual screening mammography has been recommended for asymptomatic patients. However, mammography is not without limitations. The superposition of glandular tissue is an inherent consequence of the two dimensional nature of mammography, and may result in the masking of lesions of interest (reducing sensitivity) or the false appearance of overlapping normal tissues as lesions (reducing specificity). Digital tomosynthesis of the breast is a novel x-ray based imaging technique being investigated to replace mammography for the early detection and diagnosis of breast cancer. Tomosynthesis is based on the concept of limited angle computed tomography, in which projections of the imaged object acquired from a limited angular range give enough information to partially reconstruct that object in three dimensions. Reports indicate that tomosynthesis reduces callback rate (increasing specificity) and increases detection rate (increasing sensitivity) compared to mammography. However, in current implementations of breast tomosynthesis, the breast undergoes the same vigorous compression as that used in mammography. It has been shown that the pain resulting from this vigorous breast compression during mammography is an important factor in women not undergoing the recommended annual screening mammography. This reduction in screening adherence results in an increase in breast cancer mortality and in a reduction in treatment options, yielding a higher number of mastectomies as opposed to breast conserving surgeries. The objective of this research project is to develop and test new image acquisition and processing techniques that will allow for the acquisition of breast tomosynthesis images with a substantial reduction in the amount of breast compression used with no loss in image quality or increase in radiation dose. For this, we will (1) develop a method to reduce the impact of x-ray scatter on image quality;(2) develop new image acquisition techniques for reduced compression tomosynthesis that result in the same radiation dose and image noise levels as standard tomosynthesis;(3) compare the image quality between standard tomosynthesis and reduced compression tomosynthesis performed with the techniques developed in this project. Successful completion of this project will result in the ability to perform breast tomosynthesis imaging with substantially reduced breast compression levels, with no increase in radiation dose, and no loss in image quality and tissue coverage. This will result in a reduction in the pain associated with breast cancer screening, resulting in an increase in screening adherence by women, which will decrease breast cancer mortality and increase the number of cases in which breast conserving surgery can be performed.

Public Health Relevance

Tomosynthesis imaging of the breast is being developed to overcome mammography's two-dimensional nature, which significantly impacts its sensitivity and specificity, hampering the early detection of breast cancer. Currently, breast tomosynthesis imaging involves the same vigorous compression of the patient's breast during image acquisition as that used in mammography. In this project we will develop the methods necessary to substantially reduce the amount of breast compression used in tomosynthesis, resulting in a substantial reduction in the pain associated with this procedure with no loss in image quality or increase in radiation dose to the breast. This will result in a higher breast cancer screening adherence, resulting in a higher early detection rate, increasing both breast cancer survival and breast conserving surgery rates.

National Institute of Health (NIH)
National Cancer Institute (NCI)
Research Project (R01)
Project #
Application #
Study Section
Special Emphasis Panel (ZRG1-BMIT-J (01))
Program Officer
Baker, Houston
Project Start
Project End
Budget Start
Budget End
Support Year
Fiscal Year
Total Cost
Indirect Cost
Emory University
Schools of Medicine
United States
Zip Code
Dance, David R; Sechopoulos, Ioannis (2016) Dosimetry in x-ray-based breast imaging. Phys Med Biol 61:R271-R304
Ramamurthy, Senthil; D'Orsi, Carl J; Sechopoulos, Ioannis (2016) X-ray scatter correction method for dedicated breast computed tomography: improvements and initial patient testing. Phys Med Biol 61:1116-35
Svahn, T M; Houssami, N; Sechopoulos, I et al. (2015) Review of radiation dose estimates in digital breast tomosynthesis relative to those in two-view full-field digital mammography. Breast 24:93-9
Pike, Robert; Sechopoulos, Ioannis; Fei, Baowei (2015) A minimum spanning forest based classification method for dedicated breast CT images. Med Phys 42:6190-202
Qin, Xulei; Lu, Guolan; Sechopoulos, Ioannis et al. (2014) Breast Tissue Classification in Digital Tomosynthesis Images Based on Global Gradient Minimization and Texture Features. Proc SPIE Int Soc Opt Eng 9034:90341V
Malliori, A; Bliznakova, K; Sechopoulos, I et al. (2014) Breast tomosynthesis with monochromatic beams: a feasibility study using Monte Carlo simulations. Phys Med Biol 59:4681-96
Feng, Steve Si Jia; D'Orsi, Carl J; Newell, Mary S et al. (2014) X-ray scatter correction in breast tomosynthesis with a precomputed scatter map library. Med Phys 41:031912
Sechopoulos, Ioannis; Bliznakova, Kristina; Fei, Baowei (2013) Power spectrum analysis of the x-ray scatter signal in mammography and breast tomosynthesis projections. Med Phys 40:101905
Sechopoulos, Ioannis (2013) A review of breast tomosynthesis. Part II. Image reconstruction, processing and analysis, and advanced applications. Med Phys 40:014302
Sechopoulos, Ioannis (2013) A review of breast tomosynthesis. Part I. The image acquisition process. Med Phys 40:014301

Showing the most recent 10 out of 14 publications