Dictionary Learning with Segmentation for Compressed-Sensing Magnetic Resonance Imaging

doi:10.11938/cjmr20160405

Abstract

Abstract: Dictionary learning (DL) builds a set of basis functions from the input data, such that the data can be represented more sparsely. Based on the fact that certain magnetic resonance (MR) images can be easily segmented, we propose an algorithm named dictionary learning with segmentation (DLS). The algorithm achieves better image reconstruction quality by optimizing construction of the dictionary and to making representation of the MR images sparser though incorporating image segmentation into dictionary learning. The experimental results on simulated datasets and in vivo images demonstrated that the proposed algorithm can yield better reconstruction relative to the traditional dictionary learning algorithm.

Key words: segmentation, compressed sensing, dictionary learning, magnetic resonance imaging(MRI)

CLC Number:

O482.53

SONG Yang, XIE Hai-bin, YANG Guang. Dictionary Learning with Segmentation for Compressed-Sensing Magnetic Resonance Imaging[J]. Chinese Journal of Magnetic Resonance, 2016, 33(4): 559-569.

References

[1] Pruessmann K P. Encoding and reconstruction in parallel MRI[J]. NMR Biomed, 2006, 19(3):288-299.
[2] McGibney G, Smith M R, Nichols S T, et al. Quantitative evaluation of several partial Fourier reconstruction algorithms used in MRI[J]. Magn Reson Med, 1993, 30(1):51-59.
[3] Donoho D L. Compressed sensing[J]. IEEE T Inform Theory, 2006, 52(4):1289-1306.
[4] Lustig M, Donoho D, Pauly J M. Sparse MRI:The application of compressed sensing for rapid MR imaging[J]. Magn Reson Med, 2007, 58(6):1182-1195.
[5] Ravishankar S, Bresler Y. MR image reconstruction from highly undersampled k-space data by dictionary learning[J]. IEEE T Med Imaging, 2011, 30(5):1028-1041.
[6] Toši? I, Frossard P. Dictionary learning[J]. IEEE Signal Proc Mag, 2011, 28(2):27-38.
[7] Griswold M A, Jakob P M, Heidemann R M, et al. Generalized autocalibrating partially parallel acquisitions (GRAPPA)[J]. Magn Reson Med, 2002, 47(6):1202-1210.
[8] Pruessmann K P, Weiger M, Scheidegger M B, et al. SENSE:sensitivity encoding for fast MRI[J]. Magn Reson Med, 1999, 42(5):952-962.
[9] Ma S, Yin W, Zhang Y, et al. Computer Vision and Pattern Recognition[C]. Anchorage Alarke:IEEE, 2008.
[10] Qu X B, Zhang W R, Guo D, et al. Iterative thresholding compressed sensing MRI based on contourlet transform[J]. Inverse Prob Sci Eng, 2010, 18(6):737-758.
[11] Zheng Hui(郑慧), Han Ming-yue(韩明月), Hu Bing-wei(胡炳文), et al. Comparison of different sampling schemes in compressed sensing reconstruction for DQ-SQ experiments(不同采样模式的固体DQ-SQ实验的压缩感应重建比较)[J]. Chinese J Magn Reson(波谱学杂志), 2014, 31(4):535-547.
[12] Han Ming-yue(韩明月), Zheng Hui(郑慧), Hu Bing-wei(胡炳文), et al. Compressed sensing reconstruction with iterative soft thresholding for two-dimension solid state NMR spectra with broad peaks(迭代软阈值法压缩感知重建谱峰较宽的二维固体谱)[J]. Chinese J Magn Reson(波谱学杂志), 2015, 32(4):551-562.
[13] Kim Y, Altbach M I, Trouard T P, et al. Compressed sensing using dual-tree complex wavelet transform[C]. Proc ISMRM, 2009.
[14] Aharon M, Elad M, Bruckstein A. K-SVD:An algorithm for designing overcompletedictionaries for sparse representation[J]. IEEE T Signal Process, 2006, 54(11):4311-4322.
[15] Zhan Z F, Cai J F, Guo D, et al. Fast multi-class dictionaries learning with geometrical directions in MRI reconstruction[J]. IEEE T Biomed Eng, DOI:10.1109/TBME.2015.2503756, 2015.
[16] Lai Z Y, Qu X B, Liu Y S, et al. Image reconstruction of compressed sensing MRI using graph-based redundant wavelet transform[J]. Med Imag Anal, 2016, 27:93-104.
[17] Qu X B, Guo D, Ning B D, et al. Undersampled MRI reconstruction with patch-based directional wavelets[J]. Magn Reson Imag, 2012, 30(7):964-977.
[18] Qu X B, Hou Y K, Lam F, et al. Magnetic resonance image reconstruction from undersampled measurements using a patch-based nonlocal operator[J]. Med Imag Anal, 2014, 18(6):843-856.
[19] Clarke L P, Velthuizen R P, Phuphanich S, et al. MRI:stability of three supervised segmentation techniques[J]. Magn Reson Imag, 1993, 11(1):95-106.
[20] Clarke L P, Velthuizen R P, Camacho M A, et al. MRI segmentation:methods and applications[J]. Magn Reson Imag, 1995, 13(3):343-368.
[21] Singh M, Patel P, Khosla D, et al. Segmentation of functional MRI by K-means clustering[J]. IEEE T Nucl Sci, 1996, 43(3):2030-2036.
[22] Tropp J A. Greed is good:Algorithmic results for sparse approximation[J]. IEEE T Inform Theory, 2004, 50(10):2231-2242.
[23] Hartigan J A, Wong M A. Algorithm AS 136:A k-means clustering algorithm[J]. J Appl Stat, 1979, 28(1):100-108.
[24] Brown R W, Cheng Y C N, Haacke E M, et al. Magnetic resonance imaging:physical principles and sequence design[M]. USA:John Wiley & Sons, 2014.
[25] Cocosco C A, Kollokian V, Kwan R K S, et al. BrainWeb:Online interface to a 3D MRI simulated brain database[J]. NeuroImage, 1997, 5(4):s425.
[26] Wang Z, Bovik A C, Sheikh H R, et al. Image quality assessment:from error visibility to structural similarity[J]. IEEE T Image Process, 2004, 13(4):600-612.