Spatial distribution of adipose compartments size, shape and orientation in a CT breast image of a mastectomy specimen

Development of efficient clinical breast imaging system has become a major concern in medical imaging today. Recent advancement in 3D imaging has allowed detailed analysis of breast anatomic structures. To assist with preclinical validation of breast imaging systems, high resolution anthropomorphic software phantoms have been introduced [1]. The phantoms developed at the University of Pennsylvania and Delaware State University simulate the breast outline covered by a skin layer, and the breast interior matrix of Cooper's ligaments which define compartments filled with adipose or fibro-glandular (dense) tissue. The simulation algorithms require input parameters such as number of compartments, distribution, and size and shape of adipose compartments. To obtain more realistic phantoms, these simulation parameters should be inferred from clinical images. In this study, we have investigated the distribution and spatial placement of adipose compartments in reconstructed CT images of a mastectomy specimen.