Title :
Distributed 3-D iterative reconstruction for quantitative SPECT
Author :
Ju, Z.-W. ; Frey, E.C. ; Tsui, B.M.W.
Author_Institution :
Dept. of Biomed. Eng., North Carolina Univ., Chapel Hill, NC, USA
fDate :
8/1/1995 12:00:00 AM
Abstract :
We describe a distributed three dimensional (3-D) iterative reconstruction library for quantitative single-photon emission computed tomography (SPECT). This library includes 3-D projector-backprojector pairs (PBPs) and distributed 3-D iterative reconstruction algorithms. The 3-D PBPs accurately and efficiently model various combinations of the image degrading factors including attenuation, detector response and scatter response. These PBPs were validated by comparing projection data computed using the projectors with that from direct Monte Carlo (MC) simulations. The distributed 3-D iterative algorithms spread the projection-backprojection operations for all the projection angles over a heterogeneous network of single or multi-processor computers to reduce the reconstruction time. Based on a master/slave paradigm, these distributed algorithms provide dynamic load balancing and fault tolerance. The distributed algorithms were verified by comparing images reconstructed using both the distributed and non-distributed algorithms. Computation times for distributed 3-D reconstructions running on up to 4 identical processors were reduced by a factor approximately 80-90% times the number of the processors participating, compared to those for non-distributed 3-D reconstructions running on a single processor. When combined with faster affordable computers, this library provides an efficient means for implementing accurate reconstruction and compensation methods to improve quality and quantitative accuracy in SPECT images
Keywords :
Monte Carlo methods; distributed algorithms; image reconstruction; single photon emission computed tomography; 3D projector-backprojector pairs; SPECT; attenuation; detector response; direct Monte Carlo simulations; distributed 3D iterative reconstruction; distributed algorithms; dynamic load balancing; fault tolerance; heterogeneous network; image degrading factors; master/slave paradigm; multi-processor computers; nondistributed algorithms; projection angles; quantitative SPECT; quantitative single-photon emission computed tomography; reconstruction time; scatter response; single-processor computers; Attenuation; Computed tomography; Degradation; Detectors; Distributed algorithms; Distributed computing; Image reconstruction; Libraries; Reconstruction algorithms; Three dimensional displays;
Journal_Title :
Nuclear Science, IEEE Transactions on