Meeting Banner
Abstract #1415

Composite Histogram Constrained Artifact Suppression (CHiCA) for Dynamic Cardiac Magnetic Resonance Imaging

Thomas Gaass1, Guillaume Potdevin2, Grzegorz Bauman3, 4, Peter Nol5, Axel Haase1

1Zentralinstitut fr Medizintechnik, Technische Universitt Mnchen, Garching, Germany; 2Department of Physics, Technische Universitt Mnchen, Garching, Germany; 3Division of Medical Physics in Radiology, German Cancer Research Center, Heidelberg, Germany; 4Department of Medical Physics, University of Wisconsin, Madison, WI, United States; 5Department of Diagnostic and Interventional Radiology, Technische Universitt Mnchen, Munich, Germany

We introduce a novel reconstruction technique using a composite image histogram as constraint (CHiCA) in a non-linear reconstruction framework for the suppression of aliasing artifacts in sub-Nyquist-sampled MRI. The performance of CHiCA is successfully presented on a numerical simulation of a DCE MRI measurement and a simulated radial in vivo dynamic cardiac MR. Data acquisition was implemented as interleaved radial sampling with sliding window reconstruction to generate the composite image. The reconstruction of undersampled data by constraining the images histogram offers the possibility to significantly reduce measurement time, while effectively suppressing aliasing artifacts.

Abstract PDF Poster PDF

Keywords

accurately acquisition agent aliasing alternative application applied applying around artifact artifacts assessed basis beat beginning better cancer cardiac category chamber chosen circles column composite comprising consecutive constrained constraint contrast correct covers cross cycle cycles dashed dataset denotes depict described details diagnostic domain dynamic edges effective efficiently enhanced equation equidistant especially evaluated even evolution examination fast fidelity five flight frame frames fully function generate heart hinders histogram human illustrate illustrates implemented in vivo includes inflow intensity interleaved investigations leads linear maintaining manner match measure medical mentioned minimization minor motion myocardium numerical object objects overall part perfect performance peter physics physiological portion previously proves pulmonary radial radially radiology reconstruction reconstructions recovered recovering reduction redundancies regularization remaining representative residual resolution resolved sampled scanner section sets sharpen signature simulation since sliding solid space spaces spatial spleen spokes still subsequently successfully suppression systolic take target temporal term theory third throughout transformation vessel view visibility volunteer washout white whole