PURPOSE: This work investigates the dose characteristics and image quality of a multisource cone-beam CT scanner dedicated for extremity imaging. METHODS: The scanner has an x-ray source with three separate anode-cathode units evenly distributed along the longitudinal direction. A nominal scan protocol fires the three sources sequentially, and a total of 600 projections (200 for each source) are acquired over a source-detector orbit of 210o . Dose was measured using a Farmer chamber in three CTDI phantoms stacked end-to-end. Measurements were performed at the central and four peripheral locations of a CTDI phantom on the axial plane and repeated along the longitudinal direction. The extent of 3D sampling of the three-source configuration was assessed in the Fourier domain through noise power spectrum measurements from air scans and compared with that from a single-source scan. A modified Defrise phantom and anthropomorphic knee and hand phantoms were used for visual assessment of cone-beam artifacts in the reconstructed images. RESULTS: The dose distribution for the three-source configuration exhibits radial asymmetry on the axial plane consistent with a short-scan geometry. Along the longitudinal direction, the highest dose was measured at the central axial plane where the field of view (FOV) from all three sources overlaps and falls off more slowly toward the end compared to a single-source configuration. The extent of 3D sampling is improved throughout the FOV as each source compensates for missing frequencies from the adjacent source. As a result, the reduction in streak and shading artifacts is apparent in the reconstructed images of all three phantoms. The improvement in image quality from the three-source configuration is most pronounced in joint spaces farther from the central axial plane. CONCLUSIONS: Initial assessment of the multisource scanner demonstrated the advantages over single-source designs in a compact scanner with large longitudinal FOV. The reduction in cone-beam artifact is particularly valuable for extremity imaging where high-contrast articular surfaces are present away from the central axial plane and/or throughout the FOV.
PURPOSE: This work investigates the dose characteristics and image quality of a multisource cone-beam CT scanner dedicated for extremity imaging. METHODS: The scanner has an x-ray source with three separate anode-cathode units evenly distributed along the longitudinal direction. A nominal scan protocol fires the three sources sequentially, and a total of 600 projections (200 for each source) are acquired over a source-detector orbit of 210o . Dose was measured using a Farmer chamber in three CTDI phantoms stacked end-to-end. Measurements were performed at the central and four peripheral locations of a CTDI phantom on the axial plane and repeated along the longitudinal direction. The extent of 3D sampling of the three-source configuration was assessed in the Fourier domain through noise power spectrum measurements from air scans and compared with that from a single-source scan. A modified Defrise phantom and anthropomorphic knee and hand phantoms were used for visual assessment of cone-beam artifacts in the reconstructed images. RESULTS: The dose distribution for the three-source configuration exhibits radial asymmetry on the axial plane consistent with a short-scan geometry. Along the longitudinal direction, the highest dose was measured at the central axial plane where the field of view (FOV) from all three sources overlaps and falls off more slowly toward the end compared to a single-source configuration. The extent of 3D sampling is improved throughout the FOV as each source compensates for missing frequencies from the adjacent source. As a result, the reduction in streak and shading artifacts is apparent in the reconstructed images of all three phantoms. The improvement in image quality from the three-source configuration is most pronounced in joint spaces farther from the central axial plane. CONCLUSIONS: Initial assessment of the multisource scanner demonstrated the advantages over single-source designs in a compact scanner with large longitudinal FOV. The reduction in cone-beam artifact is particularly valuable for extremity imaging where high-contrast articular surfaces are present away from the central axial plane and/or throughout the FOV.
Authors: Stephen Z Liu; Matthew Tivnan; Greg M Osgood; Jeffrey H Siewerdsen; J Webster Stayman; Wojciech Zbijewski Journal: Phys Med Biol Date: 2022-07-08 Impact factor: 4.174
Authors: Stephen Z Liu; Qian Cao; Matthew Tivnan; Steven Tilley Ii; Jeffrey H Siewerdsen; J Webster Stayman; Wojciech Zbijewski Journal: Phys Med Biol Date: 2020-12-22 Impact factor: 4.174
Authors: P Wu; N Sheth; A Sisniega; A Uneri; R Han; R Vijayan; P Vagdargi; B Kreher; H Kunze; G Kleinszig; S Vogt; S F Lo; N Theodore; J H Siewerdsen Journal: Phys Med Biol Date: 2020-08-19 Impact factor: 4.174