WE‐E‐330D‐03: Feasibility of High‐Resolution Contrast Enhanced Digital Mammography

Purpose: This study is aimed at investigating the feasibility of high‐resolution contrast enhanced digital mammography (CEDM). Method and Materials: Recent studies report certain promising aspects of contrast mammography [Jong et al., Radiology 228, 842–50, 2003] for identifying subtle lesions that might not be detectable by conventional mammography. In this study we investigate certain physical aspects of high‐resolution CEDM. The objective was to study the feasibility of high resolution and low dose CEDM with acceptable contrast characteristics. We used a prototype imager [Vedantham et al., Med Phys 31, 1462–72, 2004] that consists of a 2 × 2, CCD array. The imager was operated in a 78 μm mode by pixel binning. Computational studies with a 49 kVp, W spectrum with 0.6 mm Cu added filtration (1^st HVL: 1.9 mm of Al) indicated dose levels in the range of 0.1–0.5 mGy for a 5 cm thick, 50% glandular breast for the entire mammography exam. Theoretical modeling was performed using the parallel cascaded approach described by Cunningham and Yao [Proc. SPIE 3336, 220–30, 1998, Med Phys 28, 2020–38, 2001] for various physical conditions. In addition, experimental evaluation of the physical characteristics of the imager was conducted. Results: The resolution characteristics at 10% MTF was ∼7.8 and ∼4.2 cycles/mm and the DQE(0) estimate was ∼0.4 and ∼0.65 for 150 and 450 μm thick CsI:Tl scintillators respectively. Model results for pixel size range of 39–156 μm and CsI:Tl thickness range of 150–300 μm indicate that a 250–300 μm thick CsI scintillator with an imager pixel size of 78 μm could potentially offer a reasonable trade‐off between spatial resolution and DQE(f) characteristics. Conclusion: The results suggest that high‐resolution CEDM appears to be feasible at dose levels substantially lower then digital mammography. This research was supported in part by: NIH‐NIBIB Grant RO1‐EB002123 and the Georgia Cancer Coalition.