CINENet: deep learning-based 3D cardiac CINE MRI reconstruction with multi-coil complex-valued 4D spatio-temporal convolutions.
Journal
Scientific reports
ISSN: 2045-2322
Titre abrégé: Sci Rep
Pays: England
ID NLM: 101563288
Informations de publication
Date de publication:
13 08 2020
13 08 2020
Historique:
received:
05
04
2020
accepted:
31
07
2020
entrez:
15
8
2020
pubmed:
15
8
2020
medline:
12
1
2021
Statut:
epublish
Résumé
Cardiac CINE magnetic resonance imaging is the gold-standard for the assessment of cardiac function. Imaging accelerations have shown to enable 3D CINE with left ventricular (LV) coverage in a single breath-hold. However, 3D imaging remains limited to anisotropic resolution and long reconstruction times. Recently deep learning has shown promising results for computationally efficient reconstructions of highly accelerated 2D CINE imaging. In this work, we propose a novel 4D (3D + time) deep learning-based reconstruction network, termed 4D CINENet, for prospectively undersampled 3D Cartesian CINE imaging. CINENet is based on (3 + 1)D complex-valued spatio-temporal convolutions and multi-coil data processing. We trained and evaluated the proposed CINENet on in-house acquired 3D CINE data of 20 healthy subjects and 15 patients with suspected cardiovascular disease. The proposed CINENet network outperforms iterative reconstructions in visual image quality and contrast (+ 67% improvement). We found good agreement in LV function (bias ± 95% confidence) in terms of end-systolic volume (0 ± 3.3 ml), end-diastolic volume (- 0.4 ± 2.0 ml) and ejection fraction (0.1 ± 3.2%) compared to clinical gold-standard 2D CINE, enabling single breath-hold isotropic 3D CINE in less than 10 s scan and ~ 5 s reconstruction time.
Identifiants
pubmed: 32792507
doi: 10.1038/s41598-020-70551-8
pii: 10.1038/s41598-020-70551-8
pmc: PMC7426830
doi:
Types de publication
Journal Article
Research Support, Non-U.S. Gov't
Langues
eng
Sous-ensembles de citation
IM
Pagination
13710Subventions
Organisme : British Heart Foundation
ID : RG/20/1/34802
Pays : United Kingdom
Références
Schmidt, M. et al. Novel highly accelerated real-time CINE-MRI featuring compressed sensing with k-t regularization in comparison to TSENSE segmented and real-time Cine imaging. J. Cardiovasc. Magn. Reson.15, 36. https://doi.org/10.1186/1532-429x-15-s1-p36 (2013).
doi: 10.1186/1532-429x-15-s1-p36
Hansen, M. S., Sorensen, T. S., Arai, A. E. & Kellman, P. Retrospective reconstruction of high temporal resolution cine images from real-time MRI using iterative motion correction. Magn. Reson. Med.68, 741–750. https://doi.org/10.1002/mrm.23284 (2012).
doi: 10.1002/mrm.23284
pubmed: 22190255
Feng, L. et al. 5D whole-heart sparse MRI. Magn. Reson. Med.79, 826–838. https://doi.org/10.1002/mrm.26745 (2018).
doi: 10.1002/mrm.26745
pubmed: 28497486
Coppo, S. et al. Free-running 4D whole-heart self-navigated golden angle MRI: Initial results. Magn. Reson. Med.74, 1306–1316 (2015).
doi: 10.1002/mrm.25523
Usman, M., Ruijsink, B., Nazir, M. S., Cruz, G. & Prieto, C. Free breathing whole-heart 3D CINE MRI with self-gated Cartesian trajectory. Magn. Reson. Imaging38, 129–137. https://doi.org/10.1016/j.mri.2016.12.021 (2017).
doi: 10.1016/j.mri.2016.12.021
pubmed: 28034638
pmcid: 5375620
Vincenti, G. et al. Compressed sensing single-breath-hold CMR for fast quantification of LV function, volumes, and mass. JACC. Cardiovasc. Imaging7, 882–892. https://doi.org/10.1016/j.jcmg.2014.04.016 (2014).
doi: 10.1016/j.jcmg.2014.04.016
pubmed: 25129517
Xue, H., Kellman, P., Larocca, G., Arai, A. E. & Hansen, M. S. High spatial and temporal resolution retrospective cine cardiovascular magnetic resonance from shortened free breathing real-time acquisitions. J. Cardiovasc. Magn. Reson.15, 102–102. https://doi.org/10.1186/1532-429X-15-102 (2013).
doi: 10.1186/1532-429X-15-102
pubmed: 24228930
pmcid: 3842803
Ferrazzi, G. et al. Autocalibrated multiband CAIPIRINHA with through-time encoding: Proof of principle and application to cardiac tissue phase mapping. Magn. Reson. Med.81, 1016–1030. https://doi.org/10.1002/mrm.27460 (2019).
doi: 10.1002/mrm.27460
pubmed: 30295955
Sun, C. et al. Non-Cartesian slice-GRAPPA and slice-SPIRiT reconstruction methods for multiband spiral cardiac MRI. Magn. Reson. Med.83, 1235–1249. https://doi.org/10.1002/mrm.28002 (2020).
doi: 10.1002/mrm.28002
pubmed: 31565819
Bhatia, K. K., Price, A. N., Shi, W., Hajnal, J. V. & Rueckert, D. Super-resolution reconstruction of cardiac MRI using coupled dictionary learning. Proceedings of the IEEE International Symposium on Biomedical Imaging (ISBI), 947–950 (2014).
Odille, F. et al. Isotropic 3D cardiac cine MRI allows efficient sparse segmentation strategies based on 3D surface reconstruction. Magn. Reson. Med.79, 2665–2675. https://doi.org/10.1002/mrm.26923 (2018).
doi: 10.1002/mrm.26923
pubmed: 28971520
Kressler, B. et al. Three-dimensional cine imaging using variable-density spiral trajectories and SSFP with application to coronary artery angiography. Magn. Reson. Med.58, 535–543. https://doi.org/10.1002/mrm.21365 (2007).
doi: 10.1002/mrm.21365
pubmed: 17763360
Wetzl, J. et al. Single-breath-hold 3-D CINE imaging of the left ventricle using Cartesian sampling. Magn. Reson. Mater. Phys.31, 19–31. https://doi.org/10.1007/s10334-017-0624-1 (2018).
doi: 10.1007/s10334-017-0624-1
Barkauskas, K. J. et al. Quantification of left ventricular functional parameter values using 3D spiral bSSFP and through-time non-Cartesian GRAPPA. J. Cardiovasc. Magn. Reson.16, 65. https://doi.org/10.1186/s12968-014-0065-1 (2014).
doi: 10.1186/s12968-014-0065-1
pubmed: 25231607
pmcid: 4160541
Wech, T. et al. Whole-heart cine MRI in a single breath-hold—A compressed sensing accelerated 3D acquisition technique for assessment of cardiac function. RoFo Fortschritte auf dem Gebiete der Rontgenstrahlen und der Nuklearmedizin186, 37–41. https://doi.org/10.1055/s-0033-1350521 (2014).
doi: 10.1055/s-0033-1350521
pubmed: 23996623
Makowski, M. R. et al. Single breath-hold assessment of cardiac function using an accelerated 3D single breath-hold acquisition technique–comparison of an intravascular and extravascular contrast agent. J. Cardiovasc. Magn. Reson.14, 53. https://doi.org/10.1186/1532-429x-14-53 (2012).
doi: 10.1186/1532-429x-14-53
pubmed: 22849703
pmcid: 3438073
Küstner, T. et al. Isotropic 3D Cartesian single breath-hold CINE MRI with multi-bin patch-based low-rank reconstruction. Magn. Reson. Med. https://doi.org/10.1002/mrm.28267 (2020).
doi: 10.1002/mrm.28267
pubmed: 32250492
Deshmane, A., Gulani, V., Griswold, M. A. & Seiberlich, N. Parallel MR imaging. J. Magn. Reson. Imag.36, 55–72. https://doi.org/10.1002/jmri.23639 (2012).
doi: 10.1002/jmri.23639
Lustig, M., Donoho, D. & Pauly, J. M. Sparse MRI: The application of compressed sensing for rapid MR imaging. Magn. Reson. Med.58, 1182–1195 (2007).
doi: 10.1002/mrm.21391
Otazo, R., Candès, E. & Sodickson, D. K. Low-rank plus sparse matrix decomposition for accelerated dynamic MRI with separation of background and dynamic components. Magn. Reson. Med.73, 1125–1136. https://doi.org/10.1002/mrm.25240 (2015).
doi: 10.1002/mrm.25240
pubmed: 24760724
Ravishankar, S. & Bresler, Y. MR image reconstruction from highly undersampled k-space data by dictionary learning. IEEE Trans. Med. Imag.30, 1028–1041 (2010).
doi: 10.1109/TMI.2010.2090538
Menchón-Lara, R.-M., Simmross-Wattenberg, F., Casaseca-de-la-Higuera, P., Martín-Fernández, M. & Alberola-López, C. Reconstruction techniques for cardiac cine MRI. Insights Imag. https://doi.org/10.1186/s13244-019-0754-2 (2019).
doi: 10.1186/s13244-019-0754-2
Knoll, F. et al. Deep learning methods for parallel magnetic resonance image reconstruction. arXiv:1904.01112 (2019).
Hyun, C. M., Kim, H. P., Lee, S. M., Lee, S. & Seo, J. K. Deep learning for undersampled MRI reconstruction. Phys. Med. Biol.63, 135007 (2018).
doi: 10.1088/1361-6560/aac71a
Lin, D. J., Johnson, P. M., Knoll, F. & Lui, Y. W. Artificial intelligence for MR image reconstruction: An overview for clinicians. J. Magn. Reson. Imag. https://doi.org/10.1002/jmri.27078 (2020).
doi: 10.1002/jmri.27078
Sun, J., Li, H. & Xu, Z. Deep ADMM-Net for compressive sensing MRI. Proceedings of the Advances in neural information processing systems, 10–18 (2016).
Eo, T. et al. KIKI-net: cross-domain convolutional neural networks for reconstructing undersampled magnetic resonance images. Magn. Reson. Med.80, 2188–2201. https://doi.org/10.1002/mrm.27201 (2018).
doi: 10.1002/mrm.27201
pubmed: 29624729
Schlemper, J., Caballero, J., Hajnal, J. V., Price, A. N. & Rueckert, D. A deep cascade of convolutional neural networks for dynamic MR image reconstruction. IEEE Trans. Med. Imag.37, 491–503 (2017).
doi: 10.1109/TMI.2017.2760978
Wang, S. et al. Accelerating magnetic resonance imaging via deep learning. Proceedings of the IEEE International Symposium on Biomedical Imaging (ISBI), 514–517 (IEEE) (2016).
Han, Y., Sunwoo, L. & Ye, J. C. k-space deep learning for accelerated MRI. IEEE Trans Med Imag39, 377–386. https://doi.org/10.1109/TMI.2019.2927101 (2019).
doi: 10.1109/TMI.2019.2927101
Lee, D., Yoo, J., Tak, S. & Ye, J. C. Deep residual learning for accelerated MRI using magnitude and phase networks. IEEE Trans. Biomed. Eng.65, 1985–1995. https://doi.org/10.1109/TBME.2018.2821699 (2018).
doi: 10.1109/TBME.2018.2821699
pubmed: 29993390
Qin, C. et al. Convolutional recurrent neural networks for dynamic MR image reconstruction. IEEE Trans. Med. Imag.38, 280–290. https://doi.org/10.1109/TMI.2018.2863670 (2019).
doi: 10.1109/TMI.2018.2863670
Yang, G. et al. DAGAN: Deep de-aliasing generative adversarial networks for fast compressed sensing MRI reconstruction. IEEE Trans. Med. Imag.37, 1310–1321. https://doi.org/10.1109/TMI.2017.2785879 (2018).
doi: 10.1109/TMI.2017.2785879
Mardani, M. et al. Deep generative adversarial neural networks for compressive sensing MRI. IEEE Trans. Med. Imag.38, 167–179 (2018).
doi: 10.1109/TMI.2018.2858752
Zhu, B., Liu, J. Z., Cauley, S. F., Rosen, B. R. & Rosen, M. S. Image reconstruction by domain-transform manifold learning. Nature555, 487–492 (2018).
doi: 10.1038/nature25988
Hammernik, K. et al. Learning a variational network for reconstruction of accelerated MRI data. Magn. Reson. Med.79, 3055–3071. https://doi.org/10.1002/mrm.26977 (2018).
doi: 10.1002/mrm.26977
pubmed: 29115689
Chen, F. et al. Variable-density single-shot fast spin-echo MRI with deep learning reconstruction by using variational networks. Radiology289, 366–373 (2018).
doi: 10.1148/radiol.2018180445
Fuin, N., Bustin, A., Küstner, T., Botnar, R. & Prieto, C. A Variational Neural Network for Accelerating Free-breathing Whole-Heart Coronary MR Angiography. Proceedings of the International Society for Magnetic Resonance in Medicine (ISMRM), 478 (2019).
Cheng, J. Y., Mardani, M., Alley, M. T., Pauly, J. M. & Vasanawala, S. S. DeepSPIRiT: Generalized Parallel Imaging using Deep Convolutional Neural Networks. Proceedings of the International Society for Magnetic Resonance in Medicine (ISMRM), 570 (2018).
Hosseini, S. A. H. et al. Accelerated coronary MRI with sRAKI: A database-free self-consistent neural network k-space reconstruction for arbitrary undersampling. PLoS ONE https://doi.org/10.1371/journal.pone.0229418 (2020).
doi: 10.1371/journal.pone.0229418
pubmed: 32730260
pmcid: 7144954
Sandino, C. M., Lai, P., Vasanawala, S. S. & Cheng, J. Y. Accelerating cardiac cine MRI beyond compressed sensing using DL-ESPIRiT. arXiv preprint arXiv:1911.05845 (2019).
Sandino, C. M., Lai, P., Vasanawala, S. S. & Cheng, J. Y. DL-ESPIRiT: Improving robustness to SENSE model errors in deep learning-based reconstruction. Proceedings of the International Society for Magnetic Resonance in Medicine (ISMRM), 659 (2019).
Chen, F. et al. Data-driven self-calibration and reconstruction for non-cartesian wave-encoded single-shot fast spin echo using deep learning. J. Magn. Reson. Imag.51, 841–853. https://doi.org/10.1002/jmri.26871 (2020).
doi: 10.1002/jmri.26871
Malavé, M. O. et al. Reconstruction of undersampled 3D non-cartesian image-based navigators for coronary MRA using an unrolled deep learning model. arXiv preprint arXiv:1910.11414 (2019).
Hunter, J. D. Matplotlib: A 2D graphics environment. Comput. Sci. Eng.9, 90–95. https://doi.org/10.1109/MCSE.2007.55 (2007).
doi: 10.1109/MCSE.2007.55
Tran, D. et al. A closer look at spatiotemporal convolutions for action recognition. Proceedings of the IEEE conference on Computer Vision and Pattern Recognition, 6450–6459 (2018).
Virtue, P., Stella, X. Y. & Lustig, M. Better than real: Complex-valued neural nets for MRI fingerprinting. Proceedings of the IEEE International Conference on Image Processing (ICIP), 3953–3957 (IEEE) (2017).
Cheng, J. Y., Chen, F., Alley, M. T., Pauly, J. M. & Vasanawala, S. S. Highly scalable image reconstruction using deep neural networks with bandpass filtering. arXiv preprint arXiv:1805.03300 (2018).
Edelman, R. R., Serhal, A., Pursnani, A., Pang, J. & Koktzoglou, I. Cardiovascular cine imaging and flow evaluation using Fast Interrupted Steady-State (FISS) magnetic resonance. J. Cardiovasc. Magn. Reson.20, 12 (2018).
doi: 10.1186/s12968-018-0433-3
Bastiaansen, J. A. M. & Stuber, M. Flexible water excitation for fat-free MRI at 3T using lipid insensitive binomial off-resonant RF excitation (LIBRE) pulses. Magn. Reson. Med.79, 3007–3017. https://doi.org/10.1002/mrm.26965 (2018).
doi: 10.1002/mrm.26965
pubmed: 29159947
Küstner, T. et al. 3D Cartesian fast interrupted steady-state (FISS) imaging. Magn. Reson. Med.82, 1617–1630. https://doi.org/10.1002/mrm.27830 (2019).
doi: 10.1002/mrm.27830
pubmed: 31197881
pmcid: 6772102
Pruessmann, K. P., Weiger, M., Börnert, P. & Boesiger, P. Advances in sensitivity encoding with arbitrary k-space trajectories. Magn. Reson. Med.46, 638–651. https://doi.org/10.1002/mrm.1241 (2001).
doi: 10.1002/mrm.1241
pubmed: 11590639
Uecker, M. et al. ESPIRiT—An eigenvalue approach to autocalibrating parallel MRI: Where SENSE meets GRAPPA. Magn. Reson. Med.71, 990–1001. https://doi.org/10.1002/mrm.24751 (2014).
doi: 10.1002/mrm.24751
pubmed: 23649942
pmcid: 4142121
Ravishankar, S. & Bresler, Y. MR image reconstruction from highly undersampled k-space data by dictionary learning. IEEE Trans. Med. Imaging30, 1028–1041. https://doi.org/10.1109/TMI.2010.2090538 (2011).
doi: 10.1109/TMI.2010.2090538
pubmed: 21047708
Dedmari, M. A. et al. Complex fully convolutional neural networks for MR image reconstruction. Proceedings of the International Workshop on Machine Learning for Medical Image Reconstruction, 30–38 (Springer) (2018).
Scardapane, S., Van Vaerenbergh, S., Hussain, A. & Uncini, A. Complex-valued neural networks with nonparametric activation functions. IEEE Trans. Emerg. Topics Comput. Intell. https://doi.org/10.1109/TETCI.2018.2872600 (2018).
doi: 10.1109/TETCI.2018.2872600
Hirose, A. Complex-Valued Neural Networks: Advances and Applications 75–102 (Wiley, New York, 2013).
doi: 10.1002/9781118590072
Uecker, M. & Tamir, J. mrirecon/bart: version 0.5.00. 10.5281/zenodo.3376744 (2019).
Heiberg, E. et al. Design and validation of segment—Freely available software for cardiovascular image analysis. BMC Med. Imag. https://doi.org/10.1186/1471-2342-10-1 (2010).
doi: 10.1186/1471-2342-10-1