Susan G Komen  
I've Been Diagnosed With Breast Cancer Someone I Know Was Diagnosed Share Your Story Join Us And Stay Informed Donate To End Breast Cancer
    Home > Research & Grants > Grants Program > Research Grants > Research Grants Awarded > Abstract
    Awarded Grants
    Measurement of the X-Ray Attenuation Coefficients of Breast Tissues and Tumors Using Digital Tomosynthesis

    Scientific Abstract:
    Title: Measurement of the x-ray attenuation coefficients of breast tissues and tumors using digital tomosynthesis A. Background: Studies have shown the link between the breast tissue composition and the breast cancer risk. Methods have been developed to quantitatively measure the composition and the volume of breast tissues based on the tissue density shown on mammograms. However, the ability of these methods is limited because the three dimensional (3D) distribution of breast tissues is missing on a two dimensional (2D) mammogram. It is difficult to get accurate measurements of the tissue density and the tissue volume. Breast cancer diagnosis also suffers from tissue-overlap. Although some radiologists believe that breast cancers tend to be more attenuating than normal tissues, Jackson suggested that tumor density measured from mammograms is difficult to assess and is of limited value for the prediction of the benign or malignant nature of breast masses. An accurate quantitative measurement of attenuation coefficients of tumors may aid in both cancer risk prediction and diagnosis. Recently, digital tomosynthesis mammography has been shown to be a novel and promising 3D breast imaging technique, which provides the 3D distribution of the x-ray attenuation coefficient of the breast volume. It provides an opportunity to accurately measure the attenuation coefficient and the volume of breast tissues for cancer risk prediction, and to accurately quantify the attenuation coefficient of an individual tumor for diagnosis. B. Objectives: The objectives of the proposed research are (1) to develop tomosynthesis-based methods for measurement of attenuation coefficients of breast tissues and tumors, and (2) to test the ability to use x-ray attenuation coefficient as means to differentiate benign from malignant lesions. C. Specific Aims C.1 Develop reconstruction methods and calibration phantoms to quantify the attenuation coefficient C.2 Develop analysis tools to measure the attenuation coefficient and the volume of breast tissues and tumors C.3 Compare the attenuation coefficients of benign and malignant tumors using biopsy-proved clinical data D. Study Description D.1 Develop reconstruction methods and calibration phantoms to quantify the attenuation coefficient D.1.1 Investigate factors (kVp, scatter, beam-hardening) that affects the attenuation coefficient measurement D.1.2 Develop algorithms and calibration phantoms to accurately quantify the attenuation coefficient D.2 Develop analysis tools to measure the attenuation coefficient and the volume of breast tissues and tumors D.2.1 Develop tools for tumor segmentation, tissue classification, and volume measurement D.2.2 Develop metrics based on attenuation coefficient for intra- and inter-patient comparison D.3 Compare the attenuation coefficients of benign and malignant tumors using biopsy-proved clinical data D.3.1 Develop statistical analysis tools for comparison of clinical tomosynthesis data D.3.2 Compare the attenuation coefficients of benign and malignant tumors E. Potential Outcomes and Benefits of the Research: This research will provide tools for the accurate measurement of attenuation coefficients of breast tissues as well as tumors. This novel approach can provide information that can be used to better understand and predict the risk associated with dense breast tissues. It will also permit the actual in vivo measurement of the attenuation of benign and malignant lesions. This could greatly improve detection and diagnosis and help to develop better tools for computer-aided diagnosis (CAD) of breast cancers.

    Lay Abstract:
    Title: Measurement of the x-ray attenuation coefficients of breast tissues and tumors using digital tomosynthesis Breast tissue composition has been shown to be a predictor of breast cancer risk. Currently, methods for quantification of breast tissue composition are based on the measurement of tissue density from a mammogram, which is a two dimensional (2D) projection of a three dimensional (3D) breast volume. The accuracy of these methods is limited because there is no way of accurately quantifying the true volume of fibroglandular tissue (dense) relative to fat in a mammogram. The lack of 3D inforamtion also interferes with accurate diagnosis of breast cancer. Although radiologists observe from mammograms that cancers tend to have higher attenuation than normal tissues, they cannot isolate the attenuation of an individual tumor from that of superimposed tissues. Therefore, they cannot use attenuation difference as a potential method for differentiating benign and malignant tumors. A technique that can separate overlapping structures in breast may help in both breast cancer diagnosis and risk prediction. Tomosynthesis mammography is a new imaging technique that gives the 3D distribution of the x-ray attenuation coefficient of a breast volume. The goals of this research are to develop methods to accurately measure the attenuation coefficients of breast tissues and tumors using tomosynthesis, and to test the ability of using attenuation coefficient as means to differentiate benign from malignant lesions. To achieve the goals, we will develop reconstruction methods, calibration phantoms and analysis tools to quantify the attenuation coefficients of breast tissues and tumors, and will compare the attenuation coefficients of benign and malignant tumors using clinical tomosynthesis data that has been proved by biopsy. This research will provide methods based on tomosynthesis to accurately quantify the attenuation coefficients of breast tissues and tumors. Potentially, it can be used as an objective approach to breast cancer diagnosis and risk prediction, and can be incorporated into computer-aided diagnosis (CAD).