基于多尺度低秩模型的心脏磁共振成像方法研究

doi:10.3969/j.issn.0372-2112.2017.09.024

电子学报 ›› 2017, Vol. 45 ›› Issue (9): 2218-2224.DOI: 10.3969/j.issn.0372-2112.2017.09.024

基于多尺度低秩模型的心脏磁共振成像方法研究

蒋明峰¹, 陆雨¹, 朱志军², 徐文龙³, 汪亚明¹

1. 浙江理工大学信息学院, 浙江杭州 310018;
2. 解放军第一一七医院心血管内科, 浙江杭州 310013;
3. 中国计量学院生物医学工程系, 浙江杭州 310018

收稿日期:2016-04-26 修回日期:2016-11-22 出版日期:2017-09-25
- 作者简介:
- 蒋明峰,男,1977年生于江西丰城.现为浙江理工大学信息学院教授、硕士生导师.主要研究方向为计算机医学图像处理、生物医学信号处理.E-mail:m.jiang@zstu.edu.cn;陆雨,男,1992年生于安徽明光.现为浙江理工大学信息学院硕士研究生.主要研究方向为医疗图像处理.E-mail:adamlu1992@gmail.com;朱志军,男,1962年生于浙江奉化.现为解放军第一一七医院心血管内科主任医师.主要研究方向为心血管疾病的基础及临床研究.E-mail:zhuzj0001@163.com;徐文龙,男,徐文龙男,1972年生于浙江丽水.现为中国计量大学信息工程学院教授、硕士生导师.主要研究方向为磁共振医学成像、生物医学测量.E-mail:wenlongxu@cjlu.edu.cn;汪亚明,男,1972生于江浙江衢州,现为浙江理工大学信息学院教授、博士生导师.主要研究方向为医学图像处理、计算机视觉.E-mail:ywang@zstu.edu.cn
- 基金资助:
- 国家自然科学基金 (No.61672466，No.61672476，No.61272311）; 浙江省自然科学基金 (No.LY14F010022，No.LZ15F020004）; 浙江省科技厅公益项目 (No.2015C31075）; 浙江理工大学521人才培养计划

Multi-scale Low Rank Model Based Method for Cardiac MR Imaging Reconstruction

JIANG Ming-feng¹, LU Yu¹, ZHU Zhi-jun², XU Wen-long³, WANG Ya-ming¹

1. School of Information Science and Technology, Zhejiang Sci-tech University, Hangzhou, Zhejiang 310018, China;
2. Department of Cardiology, the 117 th Hospital of PLA, Hangzhou, Zhejiang 310013, China;
3. Department of Biomedical Engineering, China Jiliang University, Hangzhou, Zhejiang 310018, China

Received:2016-04-26 Revised:2016-11-22 Online:2017-09-25 Published:2017-09-25
- Supported by:
- National Natural Science Foundation of China (No.61672466, No.61672476, No.61272311); National Natural Science Foundation of Zhejiang Province, China (No.LY14F010022, No.LZ15F020004); Publich Welfare Program of Department of Science and Technology of Zhejiang Province (No.2015C31075); 521 Talent Training Program of Zhejiang Sci-Tech University

摘要/Abstract

摘要： 本文提出一种基于多尺度低秩模型（MSL，Multi-Scale Low rank）的磁共振成像方法，该方法将矩阵分解成多尺度的块低秩矩阵之和，并将多尺度块低秩矩阵之和的最小化作为约束条件用于磁共振成像.两种不同的心脏磁共振数据用于验证本文所提出算法重构磁共振成像的精度.实验结果表明，相比于k-t SLR（k-t Sparsity Low Rank）和L+S（Low Rank plus Sparse）方法，所提出的MSL方法具有更好的重建效果，获得更高的重构信差比（signal to error ratio），并具有更好地结构相似性，但需要更长的重构时间.

关键词: 磁共振图像重构, 多尺度低秩, 信差比

Abstract: This paper presents a multi-scale low rank based method to implement cardiac MR(Magnetic Resonance) image reconstruction,which represented a data matrix as a sum of block-wise low rank matrices with increasing scales of block sizes.And the sum of block-wise low rank matrices was used as a constraint to approach the MR image reconstruction.Two different cardiac MR datasets were used to evaluate the performance of the proposed method.Compared with the state-of-art methods,such as the k-t SLR(k-t Sparsity Low Rank) method and L+S (Low rank plus Sparse) method,the proposed MSL method can offer improved reconstruction solution in terms of higher signal to error ratio and better structural similarity index,but with longer reconstruction time.

Key words: magnetic resonance imaging reconstruction, multi-scale low rank method, signal to error ratio

中图分类号:

TP302

蒋明峰, 陆雨, 朱志军, 等. 基于多尺度低秩模型的心脏磁共振成像方法研究[J]. 电子学报, 2017, 45(9): 2218-2224.

JIANG Ming-feng, LU Yu, ZHU Zhi-jun, et al. Multi-scale Low Rank Model Based Method for Cardiac MR Imaging Reconstruction[J]. Acta Electronica Sinica, 2017, 45(9): 2218-2224.

参考文献

[1] 翁卓,谢国喜,等.基于k空间加速采集的磁共振成像技术[J].中国生物医学工程学报,2010,29(5):785-792. WENG Zhuo,Xie Guo-xi,et al.Development of fast magnetic resonance imaging techniques based on k-space accelerated collection[J].Chinese Journal of Biomedical Engineering,2010,29(5):785-792.(in Chinese)
[2] KG Hollingsworth.Reducing acquisition time in clinical MRI by data undersampling and compressed sensing reconstruction[J].Physics in Medicine & Biology,2015,60(21):297-322.
[3] Z Lai,X Qu,et al.Image reconstruction of compressed sensing MRI using graph-based redundant wavelet transform[J].Medical Image Analysis,2015,27:93-104.
[4] C Prieto,M Usman,et al.Group sparse reconstruction using intensity-based clustering[J].Magnetic Resonance in Medicine,2013,69(4):1169-1179.
[5] D Liang,EV Dibella,et al.K-t ISD:dynamic cardiac MR imaging using compressed sensing with iterative support detection[J].Magnetic Resonance in Medicine,2012,68(1):41-53.
[6] S G Lingala,E Dibella,et al.Accelerating free breathing myocardial perfusion MRI using multi coil radial k-t SLR[J].Physics in Medicine & Biology,2013,58(20):7309-7327.
[7] B Trémoulhéac,N Dikaios,et al.Dynamic MR image reconstruction-separation from undersampled (k,t)-space via low-rank plus sparse prior[J].IEEE Transactions on Medical Imaging,2014,33(8):1689-1701.
[8] L Feng,M B Srichai,et al.Highly accelerated real-time cardiac cine MRI using k-t SPARSE-SENSE[J].Magnetic Resonance in Medicine,2013,70(1):64-74.
[9] L Feng,R Grimm,et al.Golden-angle radial sparse parallel MRI:combination of compressed sensing,parallel imaging,and golden-angle radial sampling for fast and flexible dynamic volumetric MRI[J].Magnetic Resonance in Medicine,2014,72(3):707-17.
[10] 彭义刚,索津莉,等.从压缩传感到低秩矩阵恢复:理论与应用[J].自动化学报,2013,39(7):981-994. PENG Yi-gang,SUO Jin-li,et al.From compressed sensing to low-rank matrix recovery:theory and applications[J].Acta Automatica Sinica,2013,39(7):981-994.(in Chinese)
[11] 孙玉宝,吴泽彬,等.联合低秩与稀疏先验的高光谱图像压缩感知重建[J].电子学报,2014,42(11):2219-2224. SUN Yu-bao,WU Ze-bin,et al.Compressed sensing reconstruction of hyperspectral imagery jointly using low rank and sparse prior[J].Acta Automatica Sinica,2014,42(11):2219-2224.(in Chinese)
[12] S G Lingala,Y Hu,et al.Accelerated first pass cardiac perfusion MRI using improved k-t SLR[A].2011 IEEE International Symposium on Biomedical Imaging[C].USA:IEEE,2011.1280-1283.
[13] A Majumdar,RK Ward,T Aboulnasr.Non-convex algorithm for sparse and low-rank recovery:application to dynamic MRI reconstruction[J].Magnetic Resonance Imaging,2013,31(3):448-455.
[14] A G Christodoulou,H Zhang,B Zhao,et al.High-resolution cardiovascular MRI by integrating parallel imaging with low-rank and sparse modeling[J].IEEE Transactions on Bio-medical Engineering,2013,60(11):3083-3092.
[15] R Otazo,E Candès,D K Sodickson.Low-rank plus sparse matrix decomposition for accelerated dynamic MRI with separation of background and dynamic components[J].Magnetic Resonance in Medicine,2015,73(3):1125-1136.
[16] S Boyd,N Parikh,E Chu,et al.Distributed optimization and statistical learning via the alternating direction method of multipliers[J].Foundations and Trends® in Machine Learning,2011,3(1):1-122.
[17] Y Hu,M Jacob.Higher degree total variation (HDTV) regularization for image recovery[J].IEEE Trans Image Process,2012,21(5):2559-2571.
[18] H Jung,K Sung,et al.k-t FOCUSS:A general compressed sensing framework for high resolution dynamic MRI[J].Magnetic Resonance in Medicine,2009,61(1):103-116.
[19] Y Yu,J Jin,et al.Multidimensional compressed sensing MRI using tensor decomposition-based sparsifying transform[J].Plos One,2014,9(6):e98441.
[20] B Ning,X Qu,et al.Magnetic resonance image reconstruction using trained geometric directions in 2D redundant wavelets domain and non-convex optimization[J].Magnetic Resonance Imaging,2013,31(9):1611-1622.

基于多尺度低秩模型的心脏磁共振成像方法研究

Multi-scale Low Rank Model Based Method for Cardiac MR Imaging Reconstruction

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 2

编辑推荐

Metrics

[1]	周伯阳, 郭志民, 王延松, 阮伟, 吴春明, 周宁, 张伟, 程国振. 基于多尺度低秩模型的电力无线接入网异常流量检测方法[J]. 电子学报, 2020, 48(8): 1552-1557.
[2]	蒋明峰, 陆亮, 吴龙, 徐文龙, 汪亚明. 基于加权Schatten p范数最小化的磁共振图像重构方法研究[J]. 电子学报, 2019, 47(4): 784-790.