Assessment of iterative regularized parallel imaging reconstruction
Verfasser/ VerfasserinKraiger, Markus ; Clason, Christian ; Knoll, Florian ; Stollberger, Rudolf
Erschienen in
Biomedizinische Technik/Biomedical Engineering, Berlin [u.a.], 2010, Jg. 55, H. Suppl. 1, S. 234-236
Erschienende Gruyter
DokumenttypAufsatz in einer Zeitschrift
Schlagwörter (GND)Bildverarbeitung / Kernspintomografie / Online-Publikation
URNurn:nbn:at:at-ubg:3-1189 Persistent Identifier (URN)
 Das Werk ist frei verfügbar
Assessment of iterative regularized parallel imaging reconstruction [0.25 mb]
Zusammenfassung (Englisch)

Nonlinear parallel imaging reconstruction using an iterative regularized Gauss Newton method (IRGN) has shown itspotential in several applications. This technique acquires data using array coils were the coil sensitivites and the imageare both computed from the under sampled data. Several normalization steps for each image and each receiver coil arenecessary prior image reconstruction. In the current work it was examined if the complex reconstruction process limitsthe applicability of this method for quantitative determination of NMR parameters (T2). Hence measurements with fullencoding using the established parallel imaging method „Grappa“ were compared with the IRGN reconstruction. It wasshown that both parallel imaging techniques are in principle applicable for quantitative determination. In the case of theapplied cartesian image encoding certain advantages of the IRGN reconstruction in comparison to the applied "Grappa"implementation were found in areas with low SNR. Further examinations are necessary for radial k-space sampling.