Abstract Archives of the RSNA, 2004
SSC17-05
Technical Evaluation of a CT Scanner Designed for Breast Cancer Screening
Scientific Papers
Presented on November 29, 2004
Presented as part of SSC17: Physics (CT Image Quality)
John Michael Boone PhD, Presenter: Nothing to Disclose
Alexander L. C. Kwan PhD, Abstract Co-Author: Nothing to Disclose
Thomas Roy Nelson PhD, Abstract Co-Author: Nothing to Disclose
Karen K. Lindfors MD, Abstract Co-Author: Nothing to Disclose
Nikula Shah MS, Abstract Co-Author: Nothing to Disclose
James Anthony Seibert PhD, Abstract Co-Author: Nothing to Disclose
Mammography remains the primary breast cancer screening modality, however many investigators are studying the potential of other modalities for breast cancer screening in subgroups of women, including those with dense breasts, implants, and heightened risk for the disease. The goal of the Breast Tomography Project at our institution is to evaluate the potential for dedicated breast CT for cancer screening, diagnosis, and image-guided therapy.
A CT scanner tailored for imaging the female breast in the pendant position was designed and built. The prototype scanner utilizes a 30 cm x 40 cm flat panel detector and an end-windowed x-ray tube, which rotates on a gantry below the woman lying prone on a table. Acquisition time at 80 kV is currently 33 seconds, but this may be reduced to 16 seconds in the near future. The image quality was evaluated on the scanner, using realistic clinical protocols with similar dose levels to that of mammography. Spatial resolution was characterized by measuring the MTF, and low contrast test objects emulating breast tissue were used to evaluate low contrast sensitivity. The scatter to primary ratio (SPR) in the image was determined for 10, 14, and 18 cm diameter breast phantoms. The radiation dose levels were determined from a combination of Monte Carlo simulations and physical measurements.
Nyquist resolution at isocenter was determined to be 0.22 mm-2, and depending upon the reconstruction filter utilized objects on the order of 0.3 – 0.4 mm can be visualized. Measured contrast resolution is currently below theoretical levels, and this is thought to be due to the relatively high scatter-to-primary ratios (SPR’s are 0.2, 0.5, and 0.9 for 50% glandular breast phantoms measuring 10 cm, 14 cm, and 18 cm in diameter respectively).
A dedicated breast CT scanner has been designed and fabricated. The technical evaluation of the system has revealed excellent performance in some areas (spatial resolution), but has also illustrated design aspects which need improvement (scatter).
Boone, J,
Kwan, A,
Nelson, T,
Lindfors, K,
Shah, N,
Seibert, J,
Technical Evaluation of a CT Scanner Designed for Breast Cancer Screening. Radiological Society of North America 2004 Scientific Assembly and Annual Meeting, November 28 - December 3, 2004 ,Chicago IL.
http://archive.rsna.org/2004/4411641.html