[OA244] Monochromatic phase-contrast breast CT: A phantom-based image comparison with a clinical system. (August 2018)

• Record Type: Journal Article
• Authors: Fedon, Christian; Brombal, Luca; Caballo, Marco; Trapani, Vittorio Di; Donato, Sandro; Michielsen, Koen; Sechopoulos, Ioannis; Longo, Renata
• Is Part Of: Physica medica. Volume 52(2018)Supplement 1
• Page Start: 91
• Page End: 92
• Publication Date: 2018-08
• Subjects: Medical physics -- Periodicals; Biophysics -- Periodicals; Biophysics -- Periodicals; Imagerie médicale -- Périodiques; Radiothérapie -- Périodiques; Rayons X -- Sécurité -- Mesures -- Périodiques; Physique -- Périodiques; Médecine -- Périodiques; 610.153
• DOI: 10.1016/j.ejmp.2018.06.316
• Languages: English
• ISSNs: 1120-1797

Abstract

Abstract : Purpose: To compare the image quality obtained with a monochromatic phase-contrast (PhC) breast CT setup based on synchrotron radiation (SR) to that from a clinical breast CT system. Methods: A homogeneous semi-ellipsoidal breast phantom, composed of 100% adipose breast equivalent material, was used with a variety of targets (CaCO3 specks, spheroidal masses) embedded. The phantom was imaged with a clinical breast CT system (Koning Corporation, USA) based on a flat panel detector with a W/Al spectrum at 49 kVp at three different average glandular dose levels (6.5, 12.9 and 25.9 mGy). The SR-based setup consisted of a coherent, monochromatic laminar X-ray beam with a CdTe photon-counting detector placed 2 m from the sample, allowing for scatter-free PhC imaging. An X-ray energy of 30 keV was chosen to match the mean energy of the clinical system, and the dose was set to match those of the clinical system. Images were reconstructed using filtered back projection with a voxel size of (0.273 mm 3 ) for the clinical system and (0.060 mm 3 ) for the PhC breast CT. In the case of SR, images were reconstructed with and without the application of a phase-retrieval pre-processing algorithm. Contrast to noise ratio (CNR), lesion visibility and sharpness was evaluated. Results: The additional application of the phase-retrieval algorithm to the SR-based images increased the CNR by approximately 84% at 6.5 mGy. The CNR for the PhC breast CT with SR was higher than the clinical system by approximately 74% (low dose), 77% (medium dose) and 78% (high dose) in a 14.5 cm diameter phantom slab. The CaCO3 specks (size 0.130 mm) could be visualized in both images. However, as expected, lesions appeared sharper in the SR-based images, due to the smaller detector pixel size. PhC effects were not observed, since no PhC details were embedded in this phantom. Conclusions: Higher image quality was observed thanks to the use of a photon-counting detector combined with the scatter-free acquisition. Moreover, the phase-retrieval pre-processing contributes to increase the CNR. This study indicates that the PhC breast CT with SR setup reaches the image quality of a clinical breast CT system.