Cone-beam dedicated breast CT geometry consisting of an x-ray source and a large-area detector.
Inverse-geometry dedicated breast CT consisting of a large-area scanned source and a narrower detector array.
Illustrations of the in-plane (transverse) geometries of the inverse and cone-beam systems. If the IGCT source and cone-beam detector have the same in-plane extent, the coverage is nearly identical when the SID and DID are inverted in the two systems.
(a) The IGCT data as they are acquired. At each source location a small cone beam is emitted and collimated towards the detector. (b) An IGCT reverse cone-beam projection consisting of rays that connect one detector pixel to all source locations.
A flow chart detailing the steps for simulating and reconstructing IGCT and cone-beam images.
SPR profiles of simulated cone-beam projections and IGCT reverse projections for the three breast diameters. The solid lines represent the monoenergetic simulations, while the dotted lines represent the polyenergetic simulations. Published experimental cone-beam results are also plotted (Ref. 7).
Histograms comparing the SPR of IGCT and cone-beam rays that pass through the 10, 14, and diameter breast phantoms.
SPR profiles of a cone-beam system with the originally specified DID and with DID matched to the IGCT system . The object is the diameter phantom.
Detected scatter for an diameter breast phantom with and without a diameter tumor at the isocenter.
MTF of the simulated IGCT and cone-beam systems. The MTF is considerably worse than that reported in Ref. 36 because of the different SDDs and DIDs and because the MTF in this case is limited by the bandwidth of the reconstruction filter.
Reconstructed axial images of the diameter breast phantom simulated without the effects of scatter. The phantom contains IDC tumors of 0.5, 0.75, 1, and diameters. All images are windowed to display CT numbers from .
Reconstructed axial images of the diameter breast phantoms simulated with scatter. The cone-beam system was reconstructed with and without ideal scatter correction. All images are windowed from .
Central horizontal profiles through axial breast phantom images.
CNR in the absence of scatter. The error bars represent the standard deviation across five trials.
CNR of images simulated with scatter. The error bars represent the standard deviation across five trials.
Specifications of simulated IGCT and cone-beam systems.
Number of simulated photons per acquisition
Peak SPR for three breast diameters.
Article metrics loading...
Full text loading...