Flowchart for the proposed reconstruction algorithm for non-Cartesian acquisitions.
Lists
Only you see this
Abstract
<p>At every iteration, the current image estimate (shifted by the multiplier) is first transformed de-aliased with respect to the sparsity constraint Φ(·) (e.g. soft-thresholding in the wavelet domain) to generate <b>u</b><sup>(t)</sup>. Then, data consistency is enforced by a weighted average of the k-space of the thresholded signal (shifted by the multiplier) and the acquired gridded k-space (<b>G<sup>*</sup>s</b>), with weights determined by <i>β</i> and <b>K</b><i><sub>est</sub></i>, generating the new estimate <b>m</b><sup>(t)</sup>. The multiplier is also updated using <b>u</b><sup>(t)</sup> and <b>m</b><sup>(t)</sup>. The final image is generated via de-apodization of the estimate obtained at the end of the iterative process (k-space images depict one representative slice from the volume).</p>
BibTeX
@inproceedings{akcakaya2015flowchart,
author = {Mehmet Akçakaya and Seunghoon Nam and Tamer Basha and Keigo Kawaji and Tarokh Vahid and Reza Nezafat},
title = {{Flowchart for the proposed reconstruction algorithm for non-Cartesian acquisitions.}},
booktitle = {Figshare},
year = {2015},
doi = {10.1371/journal.pone.0107107.g001},
}Loading related content...