Microcalcification and Mass Detection with a Photon-Counting Si Detector for Low Dose Breast Computed Tomography
H Ding1*, H Cho1, F Masaki1, W Barber2, J Iwanczyk2, S Molloi1, (1) Department of Radiological Sciences, Unviersity of California, Irvine, CA, (2) Dx-ray Inc., Northridge, CA.TH-A-103-4 Thursday 8:00AM - 9:55AM Room: 103
Purpose: To investigate the feasibility of microcalcification and soft tissue lesion detection using an energy-resolved photon-counting detector based on Si strips for breast spectral computed tomography (CT).
Methods: Simulations and experiments were performed with a prototype spectral CT system equipped with a Si-based photon-counting detector with energy discrimination capability and 0.1 mm pixel pitch. Simulations were performed according to the realistic experimental setup. The attenuation coefficient was computed from the linear combination of the individual linear attenuation coefficient of compositional elements. The detector was simulated with intrinsic pixel pitch along with 4x4 pixel binning for microcalcification and soft tissue lesion detection, respectively. In the experiment, two cylindrical phantoms of 13 mm in diameter were constructed from polymethyl methacrylate (PMMA) and polyethylene (PE) plastic for microcalcification and soft tissue lesion detection, respectively. Five cylindrical holes were bored onto the PMMA phantom and filled with resin inserts, which contained five groups of calcium hydroxyapatite (HA) granules, ranging in sizes from 0.090 to 0.210 mm. Three holes of diameters of 1, 2, and 3 mm were bored into the PE phantom and filled with water, which represents the contrast between soft tissue lesion and adipose tissue background. The phantoms were imaged at 65 kVp with an Entrance Skin Air Kerma (ESAK) of 3 mGy. Images were also reconstructed at reduced dose levels by under sampling the full sinogram.
Results: The simulated phantom images agree well with those from the experiment. microcalcifications of approximately 0.1 mm can be observed with the investigated prototype spectral CT system at ESAK of 1 mGy. The contrast-to-noise ratio of water in PE background was improved with 4x4 binning.
Conclusion: The proposed Si strip detector is expected to offer superior image quality with the capability to detect microcalcifications and masses for low dose breast imaging.