压缩感知同步扫描重建及其采样方案的研究
收稿日期: 2015-05-19
修回日期: 2016-04-09
网络出版日期: 2016-06-05
A Synchronized Compressed Sensing Scan-Reconstruction Scheme in Magnetic Resonance Imaging
Received date: 2015-05-19
Revised date: 2016-04-09
Online published: 2016-06-05
压缩感知(compressed sensing,CS)-磁共振成像(magnetic resonance imaging,MRI)技术使用随机欠采样的k空间数据来重建图像,大大提高了成像速度.但典型的CS重建很费时,这也是CS-MRI临床应用的主要障碍之一.针对这一问题,该文提出了在扫描时同步进行CS图像重建的方案.在同步重建的过程中,可以实时显示重建图像的结果,用户可以根据图像质量来决定何时终止扫描,这样可以在节约扫描和重建时间的同时,更好地控制图像质量.由于预先无法确定最终的采样率,因此传统的变密度随机采样方法并不完全适用.该文设计了适用于同步重建过程的采样模式生成方案,同时提出了分段采样方法,把采样过程分为两个阶段,不同阶段使用不同的概率密度函数(probability densityfunction, PDF)确定待采样的相位编码行.模拟实验的结果表明,与使用单一密度函数的采样方案相比,分段采样方案能够在整个同步扫描重建过程中始终获得更好的图像.
关键词： 压缩感知(CS); 磁共振成像(MRI); 同步扫描重建; 采样模式
杨光 , 高芒 , 谢海滨 , 李智敏 , 张成秀 , 奚伟 , 姜小平 . 压缩感知同步扫描重建及其采样方案的研究[J]. 波谱学杂志, 2016 , 33(2) : 257 -268 . DOI: 10.11938/cjmr20160208
Under-sampled k-space data can be used to reconstruct high-quality images with compressed sensing (CS) algorithms, greatly improving the imaging speed. However, the traditional CS reconstruction is time-consuming, and this drawback constitutes a major obstacle to the routine clinical applications of CS-MRI. To reduce the total reconstruction time, we proposed here a synchronized CS scan-reconstruction scheme. In the scheme, reconstruction is carried out while the scan is still in process, such that the reconstructed images can be displayed in real-time, and the operator can terminate the scan as he/she wishes when the quality of the images acquired so far is deemed sufficient for his/her needs. The classic variable density random sampling strategy used for traditional CS reconstruction needs to be modified, since in this scheme the final sampling pattern remains unknown before the termination of the scan. In this paper, we developed an under-sampling strategy to meet the requirements of synchronized CS scan-reconstruction, in which different probability density functions (PDFs) are used for random sampling at different phases. Experimental results demonstrated that, compared to the single-PDF approach, a two-phase sampling strategy provided better reconstruction quality in the whole scan-reconstruction process.
[1] Chang C H, Ji J. Compressed sensing MRI with multichannel data using multicore processors[J]. Magn Reson Med, 2010, 64(4):1135-1139.
[2] Candès E J. Proceedings of the International Congress of Mathematicians[C]. Madrid:J Eur Math Soc, 2006.
[3] Donoho D. Compressed sensing[J]. IEEE T Inform Theory, 2006, 52(4):1289-1306.
[4] Tsaig Y, Donoho D L. Extensions of compressed sensing[J]. Signal Processing, 2006, 86(3):549-571.
[5] Cande's E J, Romberg J K, Tao T. Stable signal recovery from incomplete and inaccurate measurements[J]. Commun Pure Appl Math, 2006, 59(8):1207-1223.
[6] 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.
[7] Lustig M, Santors J M, Donoho D L, et al. k-t Sparse:High frame-rate dynamic MRI exploiting spatio-temporal sparsity[J]. Proc Annu Meeting ISMRM, 2006:2420.
[8] Nam S, Akcakaya M, Basha T, et al. Compressed sensing reconstruction for whole-heart imaging with 3D radial trajectories:A graphics processing unit implementation[J]. Magn Reson Med, 2013, 69(1):91-102.
[9] Candes E J, Romberg J K, Tao T. Robust uncertainty principles:Exact signal reconstruction from highly incomplete frequency information[J]. IEEE T Inform Theory, 2006, 52(2):489-509.
[10] Goldstein T, Osher S. The split Bregman methods for L1 regularized problems[J]. SIAM J Imaging Sci, 2009, 2(2):323-343.
[11] Smith D, Gore J, Yankeelov T, et al. Real-time compressive sensing MRI reconstruction using GPU computing and split Bregman methods[J]. Int J Biomed Imaging, 2012, doi:10.1155/2012/864827.
[12] Sung K, Hargreaves B A. High-frequency subband compressed sensing MRI using quadruplet sampling[J]. Magn Reson Med, 2013, 70(5):1306-1318.
[13] Tsai C M, Nishimura D G. Reduced aliasing artifacts using variable density k-space sampling trajectories[J]. Magn Reson Med, 2000, 43(3):452-458.
[14] Qu X B, Guo D, Ning B D, et al. Undersampled MRI reconstruction with patch-based directional wavelets[J]. Magn Reson Imaging, 2012, 30(7):964-977.
[15] Qu X B, Hou Y K, Lam F, et al. Magnetic resonance image reconstruction fromundersampled measurements using a patch-based nonlocal operator[J]. Med Image Anal, 2014, 18(6):843-856.
[16] Liu Q G, Wang S S, Yang K, et al. Highly undersampled magnetic resonance image reconstruction using two-level Bregman method with dictionary updating[J]. IEEE T Med Imaging, 2013, 32(7):1290-1301.
[17] Vellagoundar J, Reddy M R. Optimal k-space sampling scheme for compressive sampling MRI[J]. IECBES, 2012, doi:10.1109/IECBES.2012.6494108.
[18] Liu D D, Liang D, Liu X, et al. Under-sampling trajectory design for compressed sensing MRI[J]. IECBES, 2012, doi:10.1109/EMBC.2012.6345874.
[19] Ravishankar S, Bresler Y. Adaptive sampling design for compressed sensing MRI[J]. Conf Proc IEEE Eng Med Biol Soc, 2011, doi:10.1109/IEMBS.2011.6090639.
[20] Yang Y, Liu F, Xu W L, et al. Compressed sensing MRI via two-stage reconstruction[J]. IEEE Trans Biomed Eng, 2015, 62(1):110-118.
[21] Kutyniok G. Theory and applications of compressed sensing[J]. GAMM-Mitteilungen, 2013, 36(1):79-101.
[22] Lustig M, Donoho D L, Santos J M, et al. Compressed sensing MRI[J]. IEEE Signal Proc Mag, 2008, 25(2):72-82.
[23] Lustig M, Donoho D, Santos J M, et al. A look at how CS can improve on current imaging techniques[J]. IEEE Signal Proc Mag, 2008, 24(2):72-82.
/
| 〈 |
|
〉 |