Scatter enhanced breast CT using a mono-energetic first generation scanner: Feasibility study

Scatter enhanced computed tomography (SECT) is a technique that simultaneously reconstructs images of electron density (ρe) and the linear attenuation coefficient (μ). One of its most promising applications is the field of breast imaging where both ρe and μ have shown to be characteristic for different tissues. We simulated a SECT scanner which has shown to be able to reconstruct images of both μ and ρe. We also simulated a breast computed tomography (CT) phantom consisting of normal breast tissue background with regions of blood, adipose and glandular tissue. Carcinoma was also included in the phantom in addition to microcalcification (MCF) content and typical iodine tumor absorption. Reconstructions show that the contrast between different regions in the phantom and normal tissue range from 6.4 to 42.9 for μ and from 4.6 to 25.4 for the ρe. The ρe images are capable of discriminating between regions of MCF embedded in either adipose or glandular tissue which may not be the case in the image of μ. Noise had little impact on the primary image but reduces the quality of ρe image. The use of a Hann windowed filter however was able to improve the contrast-tonoise (CNR) from 2.3 to 2.7 with respect to reconstruction with no window. The ρe image provides complementary information that can potentially improve characterization for regions of MCF with sizes corresponding to 500 μm for a 1 mm³ voxel.