Meeting Banner
Abstract #2398

Nonlinear Inverse Reconstruction for T2 Mapping Using the Generating Function Formalism on Undersampled Cartesian Data

Tilman Johannes Sumpf1, Florian Knoll2, Jens Frahm1, Rudolf Stollberger2, Andreas Petrovic2, 3

1Biomedizinische NMR Forschungs GmbH am Max-Planck-Institut fuer biophysikalische Chemie, Goettingen, Germany; 2Institute of Medical Engineering, University of Technology, Graz, Austria; 3Ludwig Boltzmann Institute for Clinical Forensic Imaging, Graz, Austria

Quantitative evaluation of the T2 relaxation time is of high importance for diagnostic MRI. Standard T2 mapping procedures rely on the time-demanding acquisition of fully-sampled MSE datasets. Nonlinear inversion strategies allow for T2 mapping from undersampled data by exploiting a mono-exponential signal model. However, true MR data usually deviates from the idealized model which yields erroneous T2 estimations. A more accurate model of the MSE signal has been recently proposed with the generating function formalism (GF). This work evaluates the combination of the GF with a nonlinear inversion approach to allow for accurate T2 reconstructions from highly undersampled Cartesian data.

Abstract PDF Poster PDF

Keywords

acceleration accounts accurately acquisition allow analytical applying approaches artifacts balanced blocked channels chosen clearly clinical coefficients coil combination common complex conducted contain contrasts cost dampening defined degrades denotes density derivatives descent deviates deviation discarded domain duration echoes elements employing engineering errors evaluates evaluation evaluations excluded exploiting exponential extended fitting fixed forensic formalism formula fully function generating good gradient head heuristically highest highly ideal idealized illustrates implausible importance inhomogeneities initially institute inter inversion iterations knoll long magnitude mapping maps mask measures medical minimizing model modeled models mono much neglecting nonlinear partial pattern peak pixel practice presence prevent principle procedures profiles proof proposed provoke pulse quality quantitative recently reconstructed reconstruction reconstructions reduce refocusing respective sampled samples sampling scalar scaling scheme seems selecting selective sensitivities sets simply slice space spatial spin squares step strong strongly systematic tend term theory thereon traditional trans transform trio true truncation unknown usually validity variable vector violation wise yields