Meeting Banner
Abstract #2665

Quantitative Evaluation of Non-Linear Reconstruction Methods in MRI

Matthias Schloegl1, Florian Knoll1, Katharina Gruber1, Franz Ebner2, Rudolf Stollberger1

1Institute for medical Engineering, TU Graz, Graz, Austria; 2Universittsklinik fr Radiologie, Medizinische Universitt Graz, Graz, Austria

This study investigates about the capability of image metrics as objective tool for quality evaluation of non-linear reconstruction methods in the context of compressed sensing. Metric rating of reconstructions with TGV, IRGN-TV, l1-SPIRiT and CGSENSE with Cartesian, radial and random sub-sampled trajectories was compared to that of six experienced radiologists, focusing on overall image quality and recognizability of anatomy and pathology. Datasets from several body regions affected by identified pathologies were selected. Good correlations were found for metrics based upon models of the human visual system as well as for common image metrics when calculated for a region of interest.

Keywords

abdomen acceleration according account anatomy applied around artifacts attempt best brain calibration clinical common compressed concerning conclude conditions content context correlation correlations crosses database dataset datasets described details digital dots encoding engineering enlarged error evaluate evaluated evaluation evaluations even example expected expert explored field finally fitted five focus frames fully function good green highest human implemented impression improve improved index individual institute inter international investigated iterations iteratively jointly justifying knee knoll known lesion lesions letters linear lost many maps measures medical medulla metric metrics model modulated mutual noise original overall overestimation parallel pathology peak performance physics pose possibilities potential preselected processing promising pure quality quantitative radial radiologists random rapidly rating ratings reconstruction reconstructions respective rotational sampled score scores secondly seems self sensitivities sensitivity severity simulated situation sparsely spirit squared stable statistical still stopping structural studied sufficient suitable system taken telecommunication though title together trained union universal updated variation visual whole yields