# Systematic review of reconstruction techniques for accelerated quantitative MRI

## **4.1 Indirect reconstruction**

### 4.1.1 Image reconstruction

#### **4.1.1.1. Multi-contrast PI**

<table><thead><tr><th width="475">Title</th><th width="136" align="center">Link to data</th><th align="center">Link to code</th><th data-hidden></th><th data-hidden>Parameter estimation</th><th data-hidden>First Author last name </th></tr></thead><tbody><tr><td><a href="https://www.sciencedirect.com/science/article/pii/S1053811921005140?via%3Dihub">Streamlined magnetic resonance fingerprinting: Fast whole-brain coverage with deep-learning based parameter estimation</a></td><td align="center">NA</td><td align="center"><a href="https://github.com/khajehim/Streamlined-MRF">Code</a></td><td><a href="https://www.sciencedirect.com/science/article/pii/S1053811921005140?via%3Dihub">DOI</a></td><td>Learning-based estimation</td><td>Khajehim</td></tr></tbody></table>

#### 4.1.1.2. Regularized reconstruction

<table data-header-hidden><thead><tr><th width="470">Title</th><th align="center">Link to data</th><th align="center">Link to code</th><th data-hidden></th><th data-hidden>DOI</th><th data-hidden>Parameter estimation</th></tr></thead><tbody><tr><td><a href="https://ieeexplore.ieee.org/document/8076917">Incorporation of prior knowledge of signal behavior into the reconstruction to accelerate the acquisition of diffusion MRI data</a></td><td align="center">NA</td><td align="center"><a href="https://github.com/HGGM-LIM/compressedsensing-diffusion-lung-MRI">Code</a></td><td>Abascal</td><td><a href="https://ieeexplore.ieee.org/document/8076917">DOI</a></td><td>Model fitting</td></tr><tr><td><a href="https://doi.org/10.1101/2021.12.18.473283">Multi T1-weighted contrast imaging and T1 mapping with Compressed sensing FLAWS at 3T</a></td><td align="center">NA</td><td align="center"><a href="https://github.com/jerbeaumont/FLAWS-Tools">Code</a></td><td>Beaumont</td><td></td><td>Model fitting</td></tr><tr><td><a href="https://onlinelibrary.wiley.com/doi/10.1002/mrm.27694">High-dimensionality undersampled patch-based reconstruction (HD-PROST) for accelerated multi-contrast MRI</a></td><td align="center">NA</td><td align="center"><a href="https://github.com/AurelienBustin/PROST">Code</a></td><td>Bustin</td><td><a href="https://doi.org/10.1002/mrm.27694">DOI</a></td><td>Dictionary matching</td></tr><tr><td><a href="https://doi.org/10.1002/mrm.28025">Multi-shot diffusion-weighted MRI reconstruction with magnitude-based spatial-angular locally low-rank regularization (SPA-LLR)</a></td><td align="center">NA</td><td align="center"><a href="https://github.com/yuxinhu/SPA-LLR">Code</a></td><td>Hu</td><td><a href="https://doi.org/10.1002/mrm.28025">https://doi.org/10.1002/mrm.28025</a></td><td>Model fitting</td></tr><tr><td><a href="https://doi.org/10.1002/jmri.27383">Multishot Diffusion-Weighted MRI of the Breast With Multiplexed Sensitivity Encoding (MUSE) and Shot Locally Low-Rank (Shot-LLR) Reconstructions</a></td><td align="center">NA</td><td align="center"><a href="https://github.com/yuxinhu/shot-LLR">Code</a></td><td>Hu</td><td><a href="https://doi.org/10.1002/jmri.27383">https://doi.org/10.1002/jmri.27383</a></td><td>Model fitting</td></tr><tr><td><a href="https://doi.org/10.1007/s10334-019-00747-1">Accelerating cardiac diffusion tensor imaging combining local low-rank and 3D TV constraint</a></td><td align="center"><a href="https://www.cvrgrid.org/data/ex-vivo.html">Data</a></td><td align="center">NA</td><td>Huang</td><td><a href="https://doi.org/10.1007/s10334-019-00747-1">https://doi.org/10.1007/s10334-019-00747-1</a></td><td>Model fitting</td></tr><tr><td><a href="https://doi.org/10.1002/mp.13078">Low-rank magnetic resonance fingerprinting</a></td><td align="center">NA</td><td align="center"><a href="http://webee.techni-on.ac.il/Sites/People/YoninaEldar/software_det18.php">Code</a></td><td>Mazor</td><td><a href="https://doi.org/10.1002/mp.13078">https://doi.org/10.1002/mp.13078</a></td><td>Dictionary matching</td></tr><tr><td><a href="https://doi.org/10.1002/mrm.28205">Motion-corrected and high-resolution anatomically assisted (MOCHA) reconstruction of arterial spin labeling MRI</a></td><td align="center">NA</td><td align="center"><a href="https://github.com/Abolfazl-Mehranian/ASL-Recon.">Code</a></td><td>Mehranian</td><td><a href="https://doi.org/10.1002/mrm.28205">https://doi.org/10.1002/mrm.28205</a></td><td>Model fitting</td></tr><tr><td><a href="https://doi.org/10.1002/mrm.25773">Accelerating t1ρ cartilage imaging using compressed sensing with iterative locally adapted support detection and JSENSE</a></td><td align="center">NA</td><td align="center"><a href="http://bisp.kaist.ac.kr">Code</a></td><td>Zhou</td><td><a href="https://doi.org/10.1002/mrm.25773">https://doi.org/10.1002/mrm.25773</a></td><td>Model fitting</td></tr></tbody></table>

#### 4.1.1.3. Subspace constrained reconstruction

<table data-header-hidden><thead><tr><th width="482.3333333333333">Title</th><th width="125" align="center">Link to data</th><th align="center">Link to code</th><th data-hidden></th><th data-hidden></th><th data-hidden></th></tr></thead><tbody><tr><td><a href="https://doi.org/10.1002/mrm.26639">Low rank alternating direction method of multipliers reconstruction for MR fingerprinting</a></td><td align="center">NA</td><td align="center"><a href="https://bitbucket.org/asslaender/nyu_mrf_recon">Code</a></td><td>Asslander</td><td><a href="https://doi.org/10.1002/mrm.26639">https://doi.org/10.1002/mrm.26639</a></td><td>Dictionary matching</td></tr><tr><td><a href="https://doi.org/10.1002/mrm.27813">Highly accelerated multishot echo planar imaging through synergistic machine learning and joint reconstruction</a></td><td align="center"><a href="https://bit.ly/2QgBg9U">Data</a></td><td align="center"><a href="https://bit.ly/2QgBg9U">Code</a></td><td>Bilgic</td><td><a href="https://doi.org/10.1002/mrm.27813">https://doi.org/10.1002/mrm.27813</a></td><td>model fitting</td></tr><tr><td><a href="https://doi.org/10.1002/mrm.28295">Echo planar time-resolved imaging with Subspace constrained reconstruction and optimized spatiotemporal encoding</a></td><td align="center">NA</td><td align="center"><a href="https://github.com/utcsilab/t2shuffling-support">Code</a>1 <a href="https://github.com/mikgroup/phase_cycling">Code2</a></td><td>Dong</td><td><a href="https://doi.org/10.1002/mrm.28295">https://doi.org/10.1002/mrm.28295</a></td><td>Model fitting</td></tr><tr><td><a href="https://doi.org/10.1016/j.neuroimage.2021.117897">Variable flip angle echo planar time-resolved imaging (vFA-EPTI) for fast high-resolution gradient echo myelin water imaging</a></td><td align="center"><a href="https://figshare.com/articles/dataset/VFA-EPTI_Datasets/13211669">Data</a></td><td align="center"><a href="https://github.com/zijingd/VFA-EPTI">Code</a></td><td>Dong</td><td><a href="https://doi.org/10.1016/j.neuroimage.2021.117897">https://doi.org/10.1016/j.neuroimage.2021.117897</a></td><td>Model fitting</td></tr><tr><td><a href="https://doi.org/10.1088/1361-6420/ab4c9a">CoverBLIP: accelerated and scalable iterative matched-filtering for magnetic resonance fingerprint reconstruction</a></td><td align="center">NA</td><td align="center"><a href="https://github.com/mgolbabaee/CoverBLIP">Code</a></td><td>Golbabaee</td><td><a href="https://doi.org/10.1088/1361-6420/ab4c9a">https://doi.org/10.1088/1361-6420/ab4c9a</a></td><td>Dictionary matching</td></tr><tr><td><a href="https://doi.org/10.1016/j.media.2020.101945">Compressive MRI quantification using convex spatiotemporal priors and deep encoder-decoder networks</a></td><td align="center">NA</td><td align="center"><a href="https://github.com/mgolbabaee/LRTV-MRFResnet-for-MRFingerprinting">Code</a></td><td>Golbabaee</td><td><a href="https://doi.org/10.1016/j.media.2020.101945">https://doi.org/10.1016/j.media.2020.101945</a></td><td>Learning-based estimation</td></tr><tr><td><a href="https://doi.org/10.1002/mrm.26081">Acceleration of MR parameter mapping using annihilating filter-based low rank hankel matrix (ALOHA)</a></td><td align="center">NA</td><td align="center"><a href="https://bispl.weebly.com/aloha-for-mr-recon.html">Code</a></td><td>Lee</td><td><a href="https://doi.org/10.1002/mrm.26081">https://doi.org/10.1002/mrm.26081</a></td><td>Model fitting</td></tr><tr><td><a href="https://doi.org/10.1002/mrm.29244">Low-rank inversion reconstruction for through-plane accelerated radial MR fingerprinting applied to relaxometry at 0.35 T</a></td><td align="center"><a href="https://github.com/nmickevicius/mrfCaipiNLM_MRM">Data</a></td><td align="center"><a href="https://github.com/nmickevicius/mrfCaipiNLM_MRM">Code</a></td><td>Mickevicius</td><td><a href="https://doi.org/10.1002/mrm.29244">https://doi.org/10.1002/mrm.29244</a></td><td>Dictionary matching</td></tr><tr><td><a href="https://doi.org/10.1016/j.jmr.2018.03.006">Using a local low rank plus sparse reconstruction to accelerate dynamic hyperpolarized 13C imaging using the bSSFP sequence</a></td><td align="center">NA</td><td align="center"><a href="https://github.com/LarsonLab/hyperpolarized-mri-toolbox">Code</a></td><td>Milshteyn</td><td><a href="https://doi.org/10.1016/j.jmr.2018.03.006">https://doi.org/10.1016/j.jmr.2018.03.006</a></td><td>Model fitting</td></tr><tr><td><a href="https://ieeexplore.ieee.org/document/8822745">Joint T1 and T2 Mapping With Tiny Dictionaries and Subspace constrained-Constrained Reconstruction</a></td><td align="center"><a href="https://github.com/volroe/tiny-dictionaries">Data</a></td><td align="center"><a href="https://github.com/volroe/tiny-dictionaries/blob/master/main.m">Code</a></td><td>Roeloffs</td><td><a href="https://doi.org/10.1109/tmi.2019.2939130">10.1109/TMI.2019.2939130</a></td><td>Model fitting</td></tr><tr><td><a href="https://doi.org/10.1016/j.neuroimage.2022.118963">3D Echo Planar Time-resolved Imaging (3D-EPTI) for ultrafast multi-parametric quantitative MRI</a></td><td align="center"><a href="https://doi.org/10.6084/m9.figshare.14558154">Data</a></td><td align="center"><a href="https://github.com/Fuyixue/3D-EPTI.">Code</a></td><td>Wang</td><td><a href="https://doi.org/10.1016/j.neuroimage.2022.118963">https://doi.org/10.1016/j.neuroimage.2022.118963</a></td><td>Dictionary matching</td></tr><tr><td><a href="https://doi.org/10.1002/mrm.28984">An MR fingerprinting approach for quantitative inhomogeneous magnetization transfer imaging</a></td><td align="center">NA</td><td align="center"><a href="https://github.com/mriphysics/MT-MRF">Code</a></td><td>West</td><td><a href="https://doi.org/10.1002/mrm.28984">https://doi.org/10.1002/mrm.28984</a></td><td>Dictionary matching</td></tr><tr><td><a href="https://doi.org/10.1002/mrm.29219">Blip up-down acquisition for spin- and gradient-echo imaging (BUDA-SAGE) with self-supervised denoising enables efficient T2, T2*, para- and dia-magnetic susceptibility mapping</a></td><td align="center"><a href="https://github.com/zhangzijing123/BUDA-SAGE.">Data</a></td><td align="center"><a href="https://github.com/zhangzijing123/BUDA-SAGE.">Code</a></td><td>Zhang</td><td><a href="https://doi.org/10.1002/mrm.29219">https://doi.org/10.1002/mrm.29219</a></td><td>Model fitting</td></tr><tr><td><a href="https://doi.org/10.1002/mrm.29206">Cramér–Rao bound-informed training of neural networks for quantitative MRI</a></td><td align="center">NA</td><td align="center"><a href="https://github.com/quentin-duchemin/MRF-CRBLoss">Code</a></td><td>Zhang</td><td><a href="https://doi.org/10.1002/mrm.29206">https://doi.org/10.1002/mrm.29206</a></td><td>Learning-based estimation</td></tr><tr><td><a href="https://doi.org/10.1002/mrm.26701">Improved magnetic resonance fingerprinting reconstruction with low-rank and Subspace constrained modeling</a></td><td align="center">NA</td><td align="center"><a href="https://sites.google.com/site/zhaoboresearch/software">Code</a></td><td>Zhao</td><td><a href="https://doi.org/10.1002/mrm.26701">https://doi.org/10.1002/mrm.26701</a></td><td>Dictionary matching</td></tr></tbody></table>

#### 4.1.1.4. View-sharing reconstruction

<table><thead><tr><th width="486">Title</th><th align="center">Link to data</th><th align="center">Link to code</th><th data-hidden></th><th data-hidden></th><th data-hidden></th><th data-hidden></th></tr></thead><tbody><tr><td><a href="https://doi.org/10.1371/journal.pone.0201808">Accelerated magnetic resonance fingerprinting using soft-weighted key-hole (MRF-SOHO)</a></td><td align="center"><a href="https://figshare.com/articles/figure/Phantom_mat/6866075">Data</a></td><td align="center">NA</td><td>Cruz</td><td></td><td>View-sharing reconstruction</td><td>Dictionary matching</td></tr></tbody></table>

#### 4.1.1.5. Learning-based reconstruction

<table data-header-hidden><thead><tr><th width="484.3333333333333">Title</th><th width="134" align="center">Link to data</th><th align="center">Link to code</th><th data-hidden></th><th data-hidden></th><th data-hidden></th><th data-hidden></th></tr></thead><tbody><tr><td><a href="https://ieeexplore.ieee.org/document/8863423">Modl-mussels: Model-based deep learning for multishot sensitivity-encoded diffusion mri</a></td><td align="center"><a href="https://drive.google.com/open?id=10Blm-wX8ofyqLQ6w1qFcm7P-j5vcus6-">Data</a></td><td align="center"><a href="https://github.com/hkaggarwal/modl-mussels">Code</a></td><td>Aggarwal</td><td>10.1109/tmi.2019.2946501</td><td>Learning-based reconstruction</td><td>Model fitting</td></tr><tr><td><a href="https://doi.org/10.1016/j.neuroimage.2021.118404">Accelerating quantitative susceptibility and R2* mapping using incoherent undersampling and deep neural network reconstruction</a></td><td align="center">NA</td><td align="center"><a href="https://www.github.com/sunhongfu/deepMRI/tree/master/DCRNet">Code</a></td><td>Gao</td><td><a href="https://doi.org/10.1016/j.neuroimage.2021.118404">https://doi.org/10.1016/j.neuroimage.2021.118404</a></td><td>Learning-based reconstruction</td><td>Model fitting</td></tr><tr><td><a href="https://doi.org/10.1002/mrm.28446">RUN-UP: Accelerated multishot diffusion-weighted MRI reconstruction using an unrolled network with U-Net as priors</a></td><td align="center">NA</td><td align="center"><a href="https://github.com/yuxinhu/RUN-UP-for-ms-DWI">Code</a></td><td>Hu</td><td><a href="https://doi.org/10.1002/mrm.28446">https://doi.org/10.1002/mrm.28446</a></td><td>Learning-based reconstruction</td><td>Model fitting</td></tr></tbody></table>

### 4.1.2 Parameter estimation

#### 4.1.2.1. Dictionary matching

<table><thead><tr><th width="494.3333333333333">Title</th><th width="130" align="center">Link to data</th><th align="center">Link to code</th><th data-hidden></th><th data-hidden></th><th data-hidden></th></tr></thead><tbody><tr><td><a href="https://doi.org/10.1002/mrm.26639">Low rank alternating direction method of multipliers reconstruction for MR fingerprinting</a></td><td align="center">NA</td><td align="center"><a href="https://bitbucket.org/asslaender/nyu_mrf_recon">Code</a></td><td>Asslander</td><td><a href="https://doi.org/10.1002/mrm.26639">https://doi.org/10.1002/mrm.26639</a></td><td>Subspace constrained reconstruction</td></tr><tr><td><a href="https://doi.org/10.1002/mrm.25156">Rapid and accurate T2 mapping from multi–spin‐echo data using Bloch‐simulation‐based reconstruction</a></td><td align="center">NA</td><td align="center"><a href="https://cai2r.net/resources/software/emc-based-t2-mapping-package">Code</a></td><td>Ben-Eliezer</td><td><a href="https://doi.org/10.1002/mrm.25156">https://doi.org/10.1002/mrm.25156</a></td><td>FTc</td></tr><tr><td><a href="https://doi.org/10.1002/mrm.27694">High-dimensionality undersampled patch-based reconstruction (HD-PROST) for accelerated multi-contrast MRI</a></td><td align="center">NA</td><td align="center"><a href="https://github.com/AurelienBustin/PROST">Code</a></td><td>Bustin</td><td><a href="https://doi.org/10.1002/mrm.27694">https://doi.org/10.1002/mrm.27694</a></td><td>Regularized reconstruction</td></tr><tr><td><a href="https://doi.org/10.1371/journal.pone.0201808">Accelerated magnetic resonance fingerprinting using soft-weighted key-hole (MRF-SOHO)</a></td><td align="center"><a href="https://figshare.com/articles/figure/Phantom_mat/6866075">Data</a></td><td align="center">NA</td><td>Cruz</td><td><a href="https://doi.org/10.1371/journal.pone.0201808">https://doi.org/10.1371/journal.pone.0201808</a></td><td>View-sharing reconstruction</td></tr><tr><td><a href="https://doi.org/10.1088/1361-6420/ab4c9a">CoverBLIP: accelerated and scalable iterative matched-filtering for magnetic resonance fingerprint reconstruction</a></td><td align="center">NA</td><td align="center"><a href="https://github.com/mgolbabaee/CoverBLIP">Code</a></td><td>Golbabaee</td><td><a href="https://doi.org/10.1088/1361-6420/ab4c9a">https://doi.org/10.1088/1361-6420/ab4c9a</a></td><td>Subspace constrained reconstruction</td></tr><tr><td><a href="https://doi.org/10.1016/j.mri.2017.12.015">Phase unwinding for dictionary compression with multiple channel transmission in magnetic resonance fingerprinting</a></td><td align="center">NA</td><td align="center"><a href="https://bitbucket.org/macloos/pnp-mrf/wiki/Home">Code</a></td><td>Lattanzi</td><td><a href="https://doi.org/10.1016/j.mri.2017.12.015">https://doi.org/10.1016/j.mri.2017.12.015</a></td><td>FTc</td></tr><tr><td><a href="https://doi.org/10.1002/mp.13078">Low-rank magnetic resonance fingerprinting</a></td><td align="center">NA</td><td align="center"><a href="http://webee.techni-on.ac.il/Sites/People/YoninaEldar/software_det18.php">Code</a></td><td>Mazor</td><td><a href="https://doi.org/10.1002/mp.13078">https://doi.org/10.1002/mp.13078</a></td><td>Regularized reconstruction</td></tr><tr><td><a href="https://doi.org/10.1002/mrm.29244">Low-rank inversion reconstruction for through-plane accelerated radial MR fingerprinting applied to relaxometry at 0.35 T</a></td><td align="center"><a href="https://github.com/nmickevicius/mrfCaipiNLM_MRM">Data</a></td><td align="center"><a href="https://github.com/nmickevicius/mrfCaipiNLM_MRM">Code</a></td><td>Mickevicius</td><td><a href="https://doi.org/10.1002/mrm.29244">https://doi.org/10.1002/mrm.29244</a></td><td>Subspace constrained reconstruction</td></tr><tr><td><a href="https://doi.org/10.1002/mrm.27947">Fast multi-component analysis using a joint sparsity constraint for MR fingerprinting</a></td><td align="center">NA</td><td align="center"><a href="https://github.com/MNagtegaal/SPIJN">Code</a></td><td>Nagtegaal</td><td><a href="https://doi.org/10.1002/mrm.27947">https://doi.org/10.1002/mrm.27947</a></td><td>FTc</td></tr><tr><td><a href="https://ieeexplore.ieee.org/document/8908815">An Efficient Method for Multi-Parameter Mapping in Quantitative MRI Using B-Spline Interpolation</a></td><td align="center">NA</td><td align="center"><a href="https://bitbucket.org/bigr_erasmusmc/dictionary_fitting.">Code</a></td><td>Valenberg</td><td><a href="https://doi.org/10.1109/TMI.2019.2954751">10.1109/TMI.2019.2954751</a></td><td>FTc</td></tr><tr><td><a href="https://doi.org/10.1016/j.neuroimage.2022.118963">3D Echo Planar Time-resolved Imaging (3D-EPTI) for ultrafast multi-parametric quantitative MRI</a></td><td align="center"><a href="https://doi.org/10.6084/m9.figshare.14558154">Data</a></td><td align="center"><a href="https://github.com/Fuyixue/3D-EPTI.">Code</a></td><td>Wang</td><td><a href="https://doi.org/10.1016/j.neuroimage.2022.118963">https://doi.org/10.1016/j.neuroimage.2022.118963</a></td><td>Subspace constrained reconstruction</td></tr><tr><td><a href="https://doi.org/10.1002/mrm.28984">An MR fingerprinting approach for quantitative inhomogeneous magnetization transfer imaging</a></td><td align="center">NA</td><td align="center"><a href="https://github.com/mriphysics/MT-MRF">Code</a></td><td>West</td><td><a href="https://doi.org/10.1002/mrm.28984">https://doi.org/10.1002/mrm.28984</a></td><td>Subspace constrained reconstruction</td></tr><tr><td><a href="https://doi.org/10.1002/mrm.26701">Improved magnetic resonance fingerprinting reconstruction with low-rank and Subspace constrained modeling</a></td><td align="center">NA</td><td align="center"><a href="https://sites.google.com/site/zhaoboresearch/software">Code</a></td><td>Zhao</td><td><a href="https://doi.org/10.1002/mrm.26701">https://doi.org/10.1002/mrm.26701</a></td><td>Subspace constrained reconstruction</td></tr></tbody></table>

#### 4.1.2.2. Learning-based estimation

<table><thead><tr><th width="512">Title</th><th width="121.33333333333331" align="center">Link to data</th><th>Link to code</th><th data-hidden></th><th data-hidden></th><th data-hidden></th></tr></thead><tbody><tr><td><a href="https://doi.org/10.1016/j.media.2020.101741">Spatially regularized parametric map reconstruction for fast magnetic resonance fingerprinting</a></td><td align="center">NA</td><td><a href="https://github.com/fabianbalsiger/mrf-reconstruction-media2020">Code</a></td><td>Balsiger</td><td><a href="https://doi.org/10.1016/j.media.2020.101741">https://doi.org/10.1016/j.media.2020.101741</a></td><td>FTc</td></tr><tr><td><a href="https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5909960/">Quantification of relaxation times in MR Fingerprinting using deep learning</a></td><td align="center">NA</td><td><a href="https://github.com/ZhenghanFang/MRF_DL">Code</a></td><td>Fang</td><td><a href="https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5909960/">https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5909960/</a></td><td>FTc</td></tr><tr><td><a href="https://doi.org/10.1016/j.media.2020.101945">Compressive MRI quantification using convex spatiotemporal priors and deep encoder-decoder networks</a></td><td align="center">NA</td><td><a href="https://github.com/mgolbabaee/LRTV-MRFResnet-for-MRFingerprinting">Code</a></td><td>Golbabaee</td><td><a href="https://doi.org/10.1016/j.media.2020.101945">https://doi.org/10.1016/j.media.2020.101945</a></td><td>Subspace constrained reconstruction</td></tr><tr><td><a href="https://doi.org/10.1016/j.neuroimage.2021.118237">Streamlined magnetic resonance fingerprinting: Fast whole-brain coverage with deep-learning based parameter estimation</a></td><td align="center">NA</td><td><a href="https://github.com/khajehim/Streamlined-MRF">Code</a></td><td>Khajehim</td><td><a href="https://doi.org/10.1016/j.neuroimage.2021.118237">https://doi.org/10.1016/j.neuroimage.2021.118237</a></td><td>Multi-contrast PI</td></tr><tr><td><a href="https://doi.org/10.1002/mrm.29206">Cramér–Rao bound-informed training of neural networks for quantitative MRI</a></td><td align="center">NA</td><td><a href="https://github.com/quentin-duchemin/MRF-CRBLoss">Code</a></td><td>Zhang</td><td><a href="https://doi.org/10.1002/mrm.29206">https://doi.org/10.1002/mrm.29206</a></td><td>Subspace constrained reconstruction</td></tr></tbody></table>

## 4.2 Direct reconstruction

### 4.2.1 Direct model-based reconstruction

<table><thead><tr><th width="463">Title</th><th align="center">Link to data</th><th align="center">Link to code</th><th data-hidden></th><th data-hidden></th></tr></thead><tbody><tr><td><a href="https://doi.org/10.1002/mrm.25558">Accelerated and motion-robust in vivo T2 mapping from radially undersampled data using bloch-simulation-based iterative reconstruction</a></td><td align="center">NA</td><td align="center"><a href="https://beneliezer-lab.com/software-downloads-page/">Code</a></td><td>Ben-Eliezer</td><td><a href="https://doi.org/10.1002/mrm.25558">https://doi.org/10.1002/mrm.25558</a></td></tr><tr><td><a href="https://doi.org/10.1016/j.media.2020.101690">Stochastic Gradient Langevin dynamics for joint parameterization of tracer kinetic models, input functions, and T1 relaxation-times from undersampled k-space DCE-MRI</a></td><td align="center">NA</td><td align="center"><a href="https://github.com/nikolaosdikaios/TracerKineticsDCEMRI">Code</a></td><td>Dikaios</td><td><a href="https://doi.org/10.1016/j.media.2020.101690">https://doi.org/10.1016/j.media.2020.101690</a></td></tr><tr><td><a href="https://dx.doi.org/10.1002%2Fmrm.25327">Accelerated MRI thermometry by direct estimation of temperature from undersampled k-space data</a></td><td align="center">NA</td><td align="center"><a href="http://www.vuiis.vanderbilt.edu/~grissowa/">Code</a></td><td>Gaur</td><td><a href="https://dx.doi.org/10.1002%2Fmrm.25327">https://dx.doi.org/10.1002%2Fmrm.25327</a></td></tr><tr><td><a href="https://doi.org/10.1002/mrm.26540">Direct estimation of tracer-kinetic parameter maps from highly undersampled brain dynamic contrast enhanced MRI</a></td><td align="center"><a href="http://https/github.com/usc-mrel/DCE_direct_recon">Data</a></td><td align="center"><a href="http://https/github.com/usc-mrel/DCE_direct_recon">Code</a></td><td>Guo</td><td><a href="https://doi.org/10.1002/mrm.26540">https://doi.org/10.1002/mrm.26540</a></td></tr><tr><td><a href="https://doi.org/10.3390/jimaging8060157">Embedded Quantitative MRI T1rho Mapping Using Non-Linear Primal-Dual Proximal Splitting</a></td><td align="center"><a href="https://zenodo.org/record/6477557#.Y--vFXbMK3A">Data</a></td><td align="center"><a href="https://zenodo.org/record/6477557#.Y--u9XbMK3A">Code</a></td><td>Hanhela</td><td><a href="https://doi.org/10.3390/jimaging8060157">https://doi.org/10.3390/jimaging8060157</a></td></tr><tr><td><a href="https://doi.org/10.1002/mrm.27502">Rapid T1 quantification from high resolution 3D data with model-based reconstruction</a></td><td align="center">NA</td><td align="center"><a href="https://github.com/IMTtugraz">Code</a></td><td>Maier</td><td><a href="https://doi.org/10.1002/mrm.27502">https://doi.org/10.1002/mrm.27502</a></td></tr><tr><td><a href="https://doi.org/10.1109/tmi.2014.2333370">Fast T2 mapping with improved accuracy using undersampled spin-echo MRI and model-based reconstructions with a generating function</a></td><td align="center">NA</td><td align="center"><a href="https://www.mpinat.mpg.de/en/frahm/">Code</a></td><td>Sumpf</td><td><a href="https://doi.org/10.1109/tmi.2014.2333370">https://doi.org/10.1109/tmi.2014.2333370</a></td></tr><tr><td><a href="https://doi.org/10.1371/journal.pone.0122611">Model-based acceleration of Look-Locker T1 mapping</a></td><td align="center"><a href="http://dx.doi.org/doi:10.5061/dryad.165r8.">Data</a></td><td align="center"><a href="https://datadryad.org/stash/dataset/doi:10.5061/dryad.165r8">Code</a></td><td>Tran-Gia</td><td><a href="https://doi.org/10.1371/journal.pone.0122611">https://doi.org/10.1371/journal.pone.0122611</a></td></tr><tr><td><a href="https://doi.org/10.1186/s12968-019-0570-3">Model-based myocardial T1 mapping with sparsity constraints using single-shot inversion-recovery radial FLASH cardiovascular magnetic resonance</a></td><td align="center">NA</td><td align="center"><a href="https://github.com/mrirecon/myocardialt1-mapping">Code</a></td><td>Wang</td><td><a href="https://doi.org/10.1186/s12968-019-0570-3">https://doi.org/10.1186/s12968-019-0570-3</a></td></tr><tr><td><a href="https://doi.org/10.1002/mrm.28497">Model-based reconstruction for simultaneous multi-slice T1 mapping using single-shot inversion-recovery radial FLASH</a></td><td align="center"><a href="https://doi.org/10.5281/zenodo.3969809">Data</a></td><td align="center"><a href="https://github.com/mrirecon/sms-T1-mapping">Code</a></td><td>Wang</td><td><a href="https://doi.org/10.1002/mrm.28497">https://doi.org/10.1002/mrm.28497</a></td></tr><tr><td><a href="https://doi.org/10.1002/mrm.28849">Sparse precontrast T1 mapping for high-resolution whole-brain DCE-MRI</a></td><td align="center"><a href="https://github.com/usc-mrel/SparsePreT1_DCE">Data</a></td><td align="center"><a href="https://github.com/mrirecon/sms-T1-mapping">Code</a></td><td>Zhu</td><td><a href="https://doi.org/10.1002/mrm.28849">https://doi.org/10.1002/mrm.28849</a></td></tr></tbody></table>
