Abstract
Magnetic Resonance Imaging (MRI) has long been considered to be among the gold standards of today’s diagnostic imaging. The most significant drawback of MRI is long acquisition times, prohibiting its use in standard practice for some applications. Compressed sensing (CS) proposes to subsample the k-space (the Fourier domain dual to the physical space of spatial coordinates) leading to significantly accelerated acquisition. However, the benefit of compressed sensing has not been fully exploited; most of the sampling densities obtained through CS do not produce a trajectory that obeys the stringent constraints of the MRI machine imposed in practice. Inspired by recent success of deep learning-based approaches for image reconstruction and ideas from computational imaging on learning-based design of imaging systems, we introduce 3D FLAT, a novel protocol for data-driven design of 3D non-Cartesian accelerated trajectories in MRI. Our proposal leverages the entire 3D k-space to simultaneously learn a physically feasible acquisition trajectory with a reconstruction method. Experimental results, performed as a proof-of-concept, suggest that 3D FLAT achieves higher image quality for a given readout time compared to standard trajectories such as radial, stack-of-stars, or 2D learned trajectories (trajectories that evolve only in the 2D plane while fully sampling along the third dimension). Furthermore, we demonstrate evidence supporting the significant benefit of performing MRI acquisitions using non-Cartesian 3D trajectories over 2D non-Cartesian trajectories acquired slice-wise.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
Bahadir, C.D., Dalca, A.V., Sabuncu, M.R.: Learning-based optimization of the under-sampling pattern in MRI. In: Chung, A.C.S., Gee, J.C., Yushkevich, P.A., Bao, S. (eds.) IPMI 2019. LNCS, vol. 11492, pp. 780–792. Springer, Cham (2019). https://doi.org/10.1007/978-3-030-20351-1_61
Bilgic, B., et al.: Wave-CAIPI for highly accelerated 3D imaging. Magn. Reson. Med. 73(6), 2152–2162 (2015). https://doi.org/10.1002/mrm.25347. https://onlinelibrary.wiley.com/doi/abs/10.1002/mrm.25347
Boyer, C., Chauffert, N., Ciuciu, P., Kahn, J., Weiss, P.: On the generation of sampling schemes for magnetic resonance imaging. SIAM J. Imaging Sci. 9, 2039–2072 (2016)
Candès, E.J., Romberg, J., Tao, T.: Robust uncertainty principles: exact signal reconstruction from highly incomplete frequency information. IEEE Trans. Inf. Theory 52, 489–509 (2006)
Dutt, A., Rokhlin, V.: Fast Fourier transforms for nonequispaced data. SIAM J. Sci. Comput. 14, 1368–1393 (1993)
Glover, G.H., Pauly, J.M.: Projection reconstruction techniques for reduction of motion effects in MRI. Magn. Reson. Med. 28, 275–289 (1992)
Gözcü, B., et al.: Learning-based compressive MRI. IEEE Trans. Med. Imaging 37, 1394–1406 (2018)
Hammernik, K., et al.: Learning a variational network for reconstruction of accelerated MRI data. Magn. Reson. Med. 79, 3055–3071 (2018)
Isensee, F., Jaeger, P.F., Full, P.M., Wolf, I., Engelhardt, S., Maier-Hein, K.H.: Automatic cardiac disease assessment on cine-MRI via time-series segmentation and domain specific features. In: Pop, M., et al. (eds.) STACOM 2017. LNCS, vol. 10663, pp. 120–129. Springer, Cham (2018). https://doi.org/10.1007/978-3-319-75541-0_13
Kingma, D.P., Ba, J.: Adam: a method for stochastic optimization. arXiv e-prints, December 2014
Larson, P., Gurney, P., Nishimura, D.: Anisotropic field-of-views in radial imaging. IEEE Trans. Med. Imaging 27, 47–57 (2007)
Lazarus, C., et al.: SPARKLING: variable-density k-space filling curves for accelerated T2*-weighted MRI. Magn. Reson. Med. 81, 3643–3661 (2019)
Lazarus, C., Weiss, P., Gueddari, L., Mauconduit, F., Vignaud, A., Ciuciu, P.: 3D sparkling trajectories for high-resolution T2*-weighted magnetic resonance imaging (2019)
Lustig, M., Donoho, D., Pauly, J.M.: Sparse MRI: the application of compressed sensing for rapid MR imaging. Magn. Reson. Med. Official J. Int. Soc. Magn. Reson. Med. 58, 1182–1195 (2007)
Sun, J., Li, H., Xu, Z., et al.: Deep ADMM-Net for compressive sensing MRI. In: Advances in Neural Information Processing Systems (2016)
Uecker, M., et al.: ESPIRiT—an eigenvalue approach to autocalibrating parallel MRI: where SENSE meets GRAPPA. Magn. Reson. Med. Official J. Int. Soc. Magn. Reson. Med. 71, 990–1001 (2014)
Van Essen, D., Ugurbil, K., et al.: The human connectome project: a data acquisition perspective. NeuroImage (2012). https://doi.org/10.1016/j.neuroimage.2012.02.018. Connectivity
Weiss, T., Senouf, O., Vedula, S., Michailovich, O., Zibulevsky, M., Bronstein, A.: PILOT: physics-informed learned optimal trajectories for accelerated MRI. arXiv e-prints, September 2019
Weiss, T., Vedula, S., Senouf, O., Bronstein, A., Michailovich, O., Zibulevsky, M.: Learning fast magnetic resonance imaging. arXiv e-prints arXiv:1905.09324, May 2019
Zbontar, J., et al.: fastMRI: an open dataset and benchmarks for accelerated MRI. arXiv preprint arXiv:1811.08839 (2018)
Zhang, Z., Romero, A., Muckley, M.J., Vincent, P., Yang, L., Drozdzal, M.: Reducing uncertainty in undersampled MRI reconstruction with active acquisition. arXiv preprint arXiv:1902.03051 (2019)
Wang, Z., Bovik, A.C., Sheikh, H.R., Simoncelli, E.P.: Image quality assessment: from error visibility to structural similarity. IEEE Trans. Image Processing 13, 600–612 (2004)
Çiçek, O., Abdulkadir, A., Lienkamp, S., Brox, T., Ronneberger, O.: 3D U-Net: learning dense volumetric segmentation from sparse annotation. arXiv arXiv:1606.06650 (2016)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
A Supplementary Material
A Supplementary Material
The sampling densities of the fixed trajectories and 3D FLAT are visualized above. Notice the change in density between the fixed initialization and the learned. Note that for visualization purposes only a fraction of the points are shown. Best viewed in color.
Compared above are three trajectories initialized using stack-of-stars. From left to right are: fixed trajectory; 4 shots of learned SOS 2D trajectories; 4 corresponding shots of learned SOS 3D trajectories. Note that the right most trajectory is actually 3D and that all the trajectories obey the MR physical constraints.
The space is processed in 3D, any plane can be depicted easily. Shown above are three planes of the same volume. The first row is the ground truth image, processed with the full k-space. The second was created with 3D FLAT initialized with a radial trajectory at an acceleration factor of 10.
A single radial trajectory is shown with its initialization. Notice the density at the high curvature parts of the black curve.
Depicted are reconstruction results of all trajectories over different acceleration factors. The images depict a sagittal plane of a sample volume. PSNR is calculated w.r.to the groundtruth image on the left most column.
Depicted are reconstruction results of all trajectories over different acceleration factors. The images depict a coronal plane of a sample volume. PSNR is calculated w.r.to the groundtruth image on the left most column.
Rights and permissions
Copyright information
© 2020 Springer Nature Switzerland AG
About this paper
Cite this paper
Alush-Aben, J., Ackerman-Schraier, L., Weiss, T., Vedula, S., Senouf, O., Bronstein, A. (2020). 3D FLAT: Feasible Learned Acquisition Trajectories for Accelerated MRI. In: Deeba, F., Johnson, P., Würfl, T., Ye, J.C. (eds) Machine Learning for Medical Image Reconstruction. MLMIR 2020. Lecture Notes in Computer Science(), vol 12450. Springer, Cham. https://doi.org/10.1007/978-3-030-61598-7_1
Download citation
DOI: https://doi.org/10.1007/978-3-030-61598-7_1
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-61597-0
Online ISBN: 978-3-030-61598-7
eBook Packages: Computer ScienceComputer Science (R0)
