Title :
Minimum-norm least-squares MR image reconstruction
Author :
Van de Walle, R. ; Desplanques, B. ; Van Hoey, G. ; Lemahieu, I.
Author_Institution :
Dept. of Electron. & Inf. Syst., Ghent Univ., Belgium
Abstract :
Deals with magnetic resonance image reconstruction from non-uniformly sampled data. The authors´ reconstruction method is based on a general formalism for finding the pseudoinverse of a linear mapping between two Hilbert spaces. The direct use of this general formalism implies the inversion of very large matrices, which results in numerical instabilities and extremely long calculation times. It is shown that these problems can be overcome by applying a sparse approximation of the matrices that are involved
Keywords :
biomedical MRI; image reconstruction; least mean squares methods; medical image processing; Hilbert spaces; MRI; extremely long calculation times; magnetic resonance imaging; medical diagnostic imaging; minimum-norm least-squares MR image reconstruction; nonuniformly sampled data; numerical instabilities; sparse approximation; Equations; Hilbert space; Image reconstruction; Image sampling; Information systems; Magnetic resonance; Magnetic resonance imaging; Numerical stability; Reconstruction algorithms; Sparse matrices;
Conference_Titel :
[Engineering in Medicine and Biology, 1999. 21st Annual Conference and the 1999 Annual Fall Meetring of the Biomedical Engineering Society] BMES/EMBS Conference, 1999. Proceedings of the First Joint
Conference_Location :
Atlanta, GA
Print_ISBN :
0-7803-5674-8
DOI :
10.1109/IEMBS.1999.804236
