DeliCS Preprocessed Data

<p>This data set consists of pre-processed MRI data as presented in <em>Deep Learning Initialized Compressed Sensing (Deli-CS) in Volumetric Spatio-Temporal Subspace Reconstruction&nbsp;</em>[1]. By downloading this dataset you will be able to re-create the figures presented in the paper using the code available on: <a href="http://github.com/SetsompopLab/deli-cs">https://github.com/SetsompopLab/deli-cs</a> .</p> <p>Each tarball named <strong>caseXXX_preprocessed.tar.gz</strong> contains data related to that subject:</p> <ul> <li><strong>deli_2min.npy</strong> is the DL genrated initial reconstruction.</li> <li><strong>init_adj_2min.npy</strong> is the inital gridding reconstructions.</li> <li><strong>ref_2min.npy</strong> is the reference LLR reconstruction (not initialized with deliCS).</li> <li><strong>ref_6min.npy</strong> is the reference LLR reconstruction using 6 min of MRF acquisition. This is considered gold standard - NOT AVAILABLE FOR TEST CASES 002-004, which are acquired in the clinic.</li> <li><strong>refine_2min_iters_20.npy </strong>is the reconstruction from the full proposed deliCS pipeline.</li> <li><strong>T1... .npy </strong>are T1 maps from various matching reconstructions</li> <li><strong>T2... .npy </strong>are T2 maps from various matching reconstructions</li> </ul> <p>Additionally, the tarball named <strong>bartcompare.tar.gz </strong>contains the <strong>ref_2min.npy </strong>density compensated Sigpy reconstruction along with <strong>bartrecon_2min.cfl </strong>and <strong>bartrecon_2min.hdr</strong>, which are the non-density compensated Bart reconstructions shown in figure 3 in [1].</p> <p>Furthermore, meta-data needed to process the data as presented in [1] are included. Some of the figure generation code requires the subspace basis and dictionary to perform dictionary matching on the fly. The tarball <strong>shared.tar.gz</strong> contains:</p> <ul> <li>the k-space trajectory for 2 min data (<strong>traj_grp16_inacc2.mat</strong>)</li> <li>the k-space trajectory for 6 min data (<strong>traj_grp48_inacc1.mat</strong>)</li> <li>the density compensation function for each trajectory (<strong>dcf_2min.npy</strong> and <strong>dcf_6min.npy</strong>)</li> <li>the subspace basis (<strong>phi.mat</strong>)</li> <li>the dictionary (<strong>dictionary.mat</strong>)</li> <li>a scaling factor for the deli reconstruction (<strong>deli_scaling_2min.npy</strong>)</li> </ul> <p>&nbsp;</p> <p>[1]&nbsp;Iyer S, Schauman S, Sandino C, et al.&nbsp;Deep Learning Initialized Compressed Sensing (Deli-CS) in Volumetric Spatio-Temporal Subspace Reconstruction.&nbsp;<em>BioRxiv:&nbsp;</em><a href="https://www.biorxiv.org/content/10.1101/2023.03.28.534431v1">https://www.biorxiv.org/content/10.1101/2023.03.28.534431v1</a></p>