Three-Dimensional Image Reconstruction in Radiology and Nuclear Medicine

1: Cone-Beam and New Geometries Reconstruction.- Comparison of three 3D reconstruction methods from cone-beam data.- Filtered backprojection algorithms for attenuated parallel and cone-beam projections sampled on a sphere.- A theorem on divergent projections.- An adaptative and constrained model for 3D X-ray vascular reconstruction.- Cone-beam algebraic reconstruction using edge-preserving regularization.- Eigen analysis of cone-beam scanning geometries.- Efficient sampling in 3D tomography: parallel schemes.- 2: SPECT Quantitation.- Analytical approaches for image reconstruction in 3D SPECT.- Quantitative brain SPECT in three dimensions: an analytical approach without transmission scans.- An analytical approach of compensation for non-uniform attenuation and 3D detector response in cardiac SPECT imaging.- Characteristics of reconstructed point response in three-dimensional spatially variant detector response compensation in SPECT.- Evaluation of fully 3D iterative scatter compensation and post-reconstruction filtering in SPECT.- An investigation of two approximation methods for improving the speed of 3D iterative reconstruction-based scatter compensation.- 3: Patient Motion and Gated SPECT.- Evaluation of a 3D OS-EM reconstruction algorithm for correction of patient motion in emission tomography.- Space-time Gibbs priors applied to gated SPECT myocardial perfusion studies.- Reconstruction of gated SPECT myocardial images images using a temporal evolution model.- 4: PET Quantitation and Reconstruction.- Design and performance of 3D single photon transmission measurement on a positron tomograph with continuously rotating detectors.- A single scatter simulation technique for scatter correction in 3D PET.- The effect of energy threshold on image variance in fully 3D PET.-FIPI: fast 3-D PET reconstruction by Fourier inversion of rebinned plane integrals.- Performance of a fast maximum likelihood algorithm for fully 3D PET reconstruction.- Author Index.