| Although x-ray projection mammography has been very effective in early detection of breast cancer, its utility is reduced in detection of small lesions in dense breasts. One drawback is the inherent superposition of parenchymal structures that makes visualization of small lesions difficult. Breast computed tomography (CT) using flat panel detectors has been developed to address this limitation by producing 3-dimensional data. Flat panels are charge integrating detectors, and therefore lack energy resolution capability. Recent advances in solid state semiconductor x-ray detector materials and associated electronics enable the investigation of x-ray imaging systems that use a photon counting, energy discriminating detector.;A computed tomography system that uses CdZnTe (CZT) photon counting detector was compared to one that uses a flat panel detector for different breast imaging tasks. The contrast-to-noise ratio improvements were 35% and 25% for the tasks of imaging iodine over polymethyl-methacrylate (PMMA) and hydroxylapatite over PMMA, respectively. Dose reductions using the CZT detector were 52.05% and 49.45% for iodine and hydroxyapatite imaging, respectively. Material decomposition is another benefit of energy resolving detectors. Using simulations, three least squares parameter estimation decomposition techniques were investigated for four-material breast imaging tasks in the image domain. Four phantoms were constructed from PMMA, polyethylene, polyoxymethylene, hydroxyapatite, and iodine. The material decomposition process was divided into segmentation and quantification tasks. All phantoms were decomposed accurately with average quantification error range of 2.76% - 15.62%. A postmortem study was also performed and found that the contrast-to-noise ratio increased by approximately 15%.;In conclusion, breast CT with CZT detectors, in comparison to flat-panel detectors, is capable of radiation dose reduction. The spectral data can be exploited to provide material specific information that is valuable to a clinician. |
