Premium
A fast convergence method with simultaneous iterative reconstruction technique for computerized tomography
Author(s) -
Yoshinaga Tetsuya
Publication year - 1999
Publication title -
international journal of imaging systems and technology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.359
H-Index - 47
eISSN - 1098-1098
pISSN - 0899-9457
DOI - 10.1002/(sici)1098-1098(1999)10:6<432::aid-ima4>3.0.co;2-i
Subject(s) - algebraic reconstruction technique , iterative reconstruction , convergence (economics) , computer science , tomography , mathematical optimization , algorithm , iterative method , dynamical systems theory , fixed point , mathematics , computer vision , physics , mathematical analysis , optics , quantum mechanics , economics , economic growth
Considering a system of simultaneous iterative reconstruction technique (SIRT) for X‐ray computerized tomography (CT) as a discrete dynamical system, the reconstruction process can be reduced to a procedure of finding a fixed point of the dynamical system. We examine a numerical method for solving fixed points of dynamical systems derived from the algebraic reconstruction technique (ART) and the expectation maximization (EM) formulation, giving rise to the very first convergence for CT reconstruction. Because the proposed method is based on the SIRT, it has an advantage for reducing metal artifact against the filtered backprojection procedure. © 2000 John Wiley & Sons, Inc. Int J Imaging Syst Technol, 10, 432–436, 1999