FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W772472362> ?p ?o ?g. }

• W772472362 endingPage "2347" @default.
• W772472362 startingPage "2347" @default.
• W772472362 abstract "Synopsis: A fast reconstruction approach for partially parallel acqusition (PPA) techniques with non-Cartesian sampling trajectories is presented in this abstract. In our approach, the regridding of the non-Cartesian sampled lines is done with a PPA reconstruction. Necessary reconstruction parameters for a shift onto a cartesian grid are derived by a SMASH-like procedure. Instead of calculating these parameters for every required shift, this calculation is performed for only a few shifts. Reconstruction parameters for intermediate shifts are interpolated resulting in dramatically reduced reconstruction times. Introduction: Non-Cartesian sampling present unique problems for parallel imaging reconstructions. In general, large linear systems are solved to reconstruct images from undersampled data (1). An alternative was proposed by Yeh et al (2), which uses non-integer shifts for a SMASH like reconstruction. This method has been extended so that more than 1 point is used in each reconstruction, however this results in a drastic increase in reconstruction time, beyond that of conjugate-gradient methods (3). In this work continuously variable density (VD) trajectories are used, similar to the VD-sampling proposed in (4), using a non-Cartesian sampling trajectory. The distance between adjacent lines in this acquisition grows continously from a minimal to a maximal value. In our approach, we use a 2 step approach. First, each line is reconstructed onto a cartesian grid position using reconstruction parameters determined by a SMASH (5) like procedure. However, instead of using a separate inversion for each required shift, the reconstruction parameters for a small number of shifts are calculated allowing the weights for intermediate shifts to be determined by interpolation, resulting in increased reconstruction speed. Missing lines are then reconstructed in a second step using integer shifts from the regridded data. Methods: For the simulations shown here, the distance between adjacent lines in the non-Cartesian VD-sampling scheme varies continiously from 0.5 to 3.0 ∆ky. Asuming a cartesian grid from -128 to +127 ∆ky, 256 phase encoding steps are necessary for the acquisition. The same range of ky can be covered with 128 VD-lines on a non-Cartesian grid, as shown in Fig. 1 in the middle. In the first step, all acquired VD-lines are regridded onto their nearest Cartesian grid position using a SMASH reconstruction with fractional harmonics. For the reconstruction, only 11 sets of reconstruction parameters were calculated, covering shifts from -∆ky to +∆ky in steps of 0.1 (see Fig.1 right, gray points). The reconstruction parameters vary slowly over the fractional harmonic values, so that all intermediate shifts necessary for the whole reconstruction can be easily interpolated from these 11 calculated sets. In a second step missing Cartesian lines in the outer part of k-space were reconstructed with the ± 1.0 ∆ky reconstruction parameters using the regridded Cartesian data lines from adjacent positions. Results: In order to demonstrate the feasibility of our approach, results from simulations are shown in Fig. 2. The direct Fourier Transformation (dFT) of a nonCartesian VD-data set with 128x256 matrix size is shown in Fig. 2a. Besides the ringing artifacts due to the low resolution, foldover artifacts are clearly visible as a result of the non uniform FOV. Application of our approach to the same data set leads to significantly reduced artifacts, see Fig. 2b. Applying the procedure in +ky and –ky direction allows one to reconstruct 256 cartesian lines from 128 VD-lines, resulting in reduced ringing artifacts. Discussion: In this work we have demonstrated that it is feasible to use interpolated reconstruction parameters for parallel imaging using a 1D non-Cartesian VD-sampling trajectory. The flexible implementation of our approach allows one to use other regenerative k-space based PPA methods, such as AUTO-SMASH (6) or VDAUTO-SMASH (7) to determine the reconstruction parameters. This technique could easliy be extended to 2D non-Cartesian trajectories (e.g. spiral) using multidimensional interpolation as well. References: (1) Yeh E, et al. ISMRM 2002 #2390 (2) Yeh E, et al. ISMRM 2001 #1796 (3) Pruessmann K, et al. Magn Reson Med 2001;46:638-651. (4) Kyriakos W, et al. Magn Reson Med 2000;44:301-308. (5) Sodickson D, et al. Magn Reson Med 1997;38:591-603. (6) Jakob P, et al. MAGMA 1998;7:42-54. (7) Heidemann R, et al. Magn Reson Med 2001;45:1066-1074. a b" @default.
• W772472362 created "2016-06-24" @default.
• W772472362 creator A5005551933 @default.
• W772472362 creator A5009828582 @default.
• W772472362 creator A5039200823 @default.
• W772472362 date "2003-01-01" @default.
• W772472362 modified "2023-09-25" @default.
• W772472362 title "Fast parallel image reconstructions with non-cartesian trajectories" @default.
• W772472362 hasPublicationYear "2003" @default.
• W772472362 type Work @default.
• W772472362 sameAs 772472362 @default.
• W772472362 citedByCount "3" @default.
• W772472362 crossrefType "journal-article" @default.
• W772472362 hasAuthorship W772472362A5005551933 @default.
• W772472362 hasAuthorship W772472362A5009828582 @default.
• W772472362 hasAuthorship W772472362A5039200823 @default.
• W772472362 hasConcept C106131492 @default.
• W772472362 hasConcept C11413529 @default.
• W772472362 hasConcept C140779682 @default.
• W772472362 hasConcept C141379421 @default.
• W772472362 hasConcept C16038011 @default.
• W772472362 hasConcept C187691185 @default.
• W772472362 hasConcept C2524010 @default.
• W772472362 hasConcept C31972630 @default.
• W772472362 hasConcept C33923547 @default.
• W772472362 hasConcept C41008148 @default.
• W772472362 hasConcept C43173174 @default.
• W772472362 hasConcept C81184566 @default.
• W772472362 hasConceptScore W772472362C106131492 @default.
• W772472362 hasConceptScore W772472362C11413529 @default.
• W772472362 hasConceptScore W772472362C140779682 @default.
• W772472362 hasConceptScore W772472362C141379421 @default.
• W772472362 hasConceptScore W772472362C16038011 @default.
• W772472362 hasConceptScore W772472362C187691185 @default.
• W772472362 hasConceptScore W772472362C2524010 @default.
• W772472362 hasConceptScore W772472362C31972630 @default.
• W772472362 hasConceptScore W772472362C33923547 @default.
• W772472362 hasConceptScore W772472362C41008148 @default.
• W772472362 hasConceptScore W772472362C43173174 @default.
• W772472362 hasConceptScore W772472362C81184566 @default.
• W772472362 hasLocation W7724723621 @default.
• W772472362 hasOpenAccess W772472362 @default.
• W772472362 hasPrimaryLocation W7724723621 @default.
• W772472362 hasRelatedWork W1507430607 @default.
• W772472362 hasRelatedWork W1549828911 @default.
• W772472362 hasRelatedWork W1986872445 @default.
• W772472362 hasRelatedWork W2016179251 @default.
• W772472362 hasRelatedWork W2041396420 @default.
• W772472362 hasRelatedWork W2041969899 @default.
• W772472362 hasRelatedWork W2082580315 @default.
• W772472362 hasRelatedWork W2111388536 @default.
• W772472362 hasRelatedWork W2114708945 @default.
• W772472362 hasRelatedWork W2147336975 @default.
• W772472362 hasRelatedWork W2151354228 @default.
• W772472362 hasRelatedWork W2160820770 @default.
• W772472362 hasRelatedWork W2182798237 @default.
• W772472362 hasRelatedWork W2374061555 @default.
• W772472362 hasRelatedWork W2511615083 @default.
• W772472362 hasRelatedWork W2618624868 @default.
• W772472362 hasRelatedWork W780051034 @default.
• W772472362 hasRelatedWork W2167233877 @default.
• W772472362 hasRelatedWork W2168887049 @default.
• W772472362 hasRelatedWork W2169670552 @default.
• W772472362 isParatext "false" @default.
• W772472362 isRetracted "false" @default.
• W772472362 magId "772472362" @default.
• W772472362 workType "article" @default.