Under previous funding, we have designed and built four breast CT scanners. The latest scanner, dubbed Doheny, makes use of a new detector system which delivers high spatial resolution and low noise breast CT images. The objectives of this proposal include software optimization and clinical trial comparisons of breast CT for screening normal risk women as well as high risk women. Software optimization: The first 2 aims of this project involve (1) the development of software which both corrects for the cupping artifacts associated with x-ray scatter, and restores the quantitative integrity of the breast CT images such that they have excellent gray-scale accuracy. A hardware-based scatter measurement procedure (reverse beam stop) has been implemented on the Doheny system;
this Specific Aim will focus on optimizing the software corrections based upon the information provided by the reverse beam stop data. (2) Due to newly available components, the Doheny scanner has 4 times the spatial resolution of our earlier systems. In this aim, we will optimize th reconstruction kernel and matrix dimensions to assure maximal performance with respect to microcalcification detection, which requires very high spatial resolution. Clinical evaluation: Fou hundred women who will be having breast biopsy as part of their standard care (BIRADS 4 and 5) will be recruited to undergo additional imaging prior to biopsy, including (research) pre-and post- contrast enhanced breast CT imaging of both breasts, as well as (standard, FDA approved) contrast enhanced breast MRI (Specific Aim 3).
In Aim 4, the potential of non-contrast enhanced breast CT will be compared using receiver operating characteristic (ROC) methodology against mammography alone, as well as with mammography + tomosynthesis.
In Specific Aim 5, the breast CT images (including both pre-and post-contrast images) will be compared using ROC methodology against standard-of-care contrast-enhanced MRI (which includes both non-contrast and contrast images). The results of the clinical trials proposed in this investigation should provide strong evidence in regards to the potential of breast CT for breast cancer screening in the normal risk and high risk populations.
Breast cancer remains as the second largest cancer killer of women, however widespread screening using mammography has reduced the mortality of this disease over the past 30 years. Nevertheless, the breast imaging community is very aware of the limitations of mammography in regards to breast cancer detection, especially in the dense breast. As a result, additional breast imaging modalities are used with increasing frequency. Breast tomosynthesis (limited angle x-ray tomography of the breast), breast ultrasound, and DCE breast MRI are all used in various settings to augment mammography. Dedicated breast CT has been studied at several sites around the world with promising results. In our laboratory, we have built 4 breast CT scanners, and we believe that our fourth breast CT scanner ('Doheny') is currently the highest resolution breast CT scanner in the world. This new (resubmitted) R01grant proposal seeks to perform software-based improvements to improve the quantitative accuracy of the breast CT images (scatter correction, etc.) and to optimize reconstruction parameters for microcalcification detection. Employing these post-acquisition software advances on our Doheny scanner, a clinical trial involving 400 BIRADS 4 and 5 women will compare, using receiver operating characteristic (ROC) analysis, non-contrast breast CT images with mammography and [mammography + tomosynthesis] for routine breast cancer screening; using the same cohort of women (their breast images), we will compare contrast enhanced breast CT with contrast enhanced MRI using ROC methodology with 4 radiologist readers to examine the potential of breast CT for high risk screening.
| Abbey, Craig K; Samuelson, Frank W; Zeng, Rongping et al. (2018) Classification images for localization performance in ramp-spectrum noise. Med Phys 45:1970-1984 |
| Kompaniez-Dunigan, Elysse; Abbey, Craig K; Boone, John M et al. (2018) Visual adaptation and the amplitude spectra of radiological images. Cogn Res Princ Implic 3:3 |
| Boone, John M; Hernandez, Andrew M; Seibert, J Anthony (2017) Two-dimensional breast dosimetry improved using three-dimensional breast image data. Radiol Phys Technol 10:129-141 |
| Hernandez, Andrew M; Seibert, J Anthony; Nosratieh, Anita et al. (2017) Generation and analysis of clinically relevant breast imaging x-ray spectra. Med Phys 44:2148-2160 |
| Lee, Juhun; Nishikawa, Robert M; Reiser, Ingrid et al. (2017) Optimal reconstruction and quantitative image features for computer-aided diagnosis tools for breast CT. Med Phys 44:1846-1856 |
| Boone, John M; McNitt-Gray, Michael F; Hernandez, Andrew M (2017) Monte Carlo Basics for Radiation Dose Assessment in Diagnostic Radiology. J Am Coll Radiol 14:793-794 |
| Hernandez, Andrew M; Boone, John M (2017) Average glandular dose coefficients for pendant-geometry breast CT using realistic breast phantoms. Med Phys 44:5096-5105 |
| Boone, John M; Hernandez, Andrew M (2017) The Effect of Iodine-based Contrast Material on Radiation Dose at CT: It's Complicated. Radiology 283:624-627 |
| Abbey, Craig K; Wu, Yirong; Burnside, Elizabeth S et al. (2016) A Utility/Cost Analysis of Breast Cancer Risk Prediction Algorithms. Proc SPIE Int Soc Opt Eng 9787: |
| Gazi, Peymon M; Aminololama-Shakeri, Shadi; Yang, Kai et al. (2016) Temporal subtraction contrast-enhanced dedicated breast CT. Phys Med Biol 61:6322-46 |
Showing the most recent 10 out of 11 publications
