基于虚拟线圈和卷积神经网络的多层同时激发图像重建

王婉婷; 苏适; 贾森; 梁栋; 王海峰

doi:10.11938/cjmr20202800

波谱学杂志 >

2020 , Vol. 37 >Issue 4: 407 - 421

DOI: https://doi.org/10.11938/cjmr20202800

研究论文

基于虚拟线圈和卷积神经网络的多层同时激发图像重建

王婉婷 ,
苏适 ,
贾森 ,
梁栋 ,
王海峰

展开

1. 保罗C. 劳特伯生物医学成像研究中心(中国科学院深圳先进技术研究院), 广东深圳 518055;
2. 中国科学院大学, 北京 100049;
3. 医学人工智能研究中心(中国科学院深圳先进技术研究院), 广东深圳 518055

收稿日期: 2020-01-14

网络出版日期: 2020-04-16

基金资助

深圳市孔雀团队资助项目（KQTD20180413181834876）；中国科学院B类先导专项资助项目（XDB25000000）；中国科学院影像技术和装备工程实验室资助项目（KFJ-PTXM-012）；广东省珠江人才计划资助项目（2019QN01Y986）；广东省自然科学基金资助项目（2018A0303130132）；国家自然科学基金资助项目（61771463，81830056，U1805261，81971611，61871373，81729003，81901736）.

收起

Reconstruction of Simultaneous Multi-Slice MRI Data by Combining Virtual Conjugate Coil Technology and Convolutional Neural Network

WANG Wan-ting ,
SU Shi ,
JIA Sen ,
LIANG Dong ,
WANG Hai-feng

Expand

1. Paul C. Lauterbur Research Center for Biomedical Imaging, Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences, Shenzhen 518055, China;
2. University of Chinese Academy of Sciences, Beijing 100049, China;
3. Research Center for Medical AI, Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences, Shenzhen 518055, China

Received date: 2020-01-14

Online published: 2020-04-16

Fold

摘要

本文提出一种基于虚拟共轭线圈（Virtual Coil Concept，VCC）技术和k空间插值鲁棒人工神经网络（Robust Artificial-neural-networks for k-space Interpolation，RAKI）的图像重建方法，用于磁共振多层同时激发成像（Simultaneous Multi-Slice imaging，SMS），该方法能够有效提升重建图像的质量，被命名为VIRGINIA（VIRtual conjuGate coIls Neural-networks InterpolAtion）.为了得到更高质量的SMS图像，本文提出的VIRGINIA方法利用磁共振线圈数据的复数共轭对称性质扩展了SMS所获取的多通道数据，并将扩展后的数据用于RAKI网络的训练，利用训练后的网络实现高质量的SMS图像重建.本文将VIRGINIA方法和其他SMS图像重建方法（RAKI和Slice-GRAPPA方法）进行了对比，并采用结构相似指数（Structural Similarity Index，SSIM）、峰值信噪比（Peak Signal-to-Noise Ratio，PSNR）和均方根误差（Root Mean Square Error，RMSE）对不同方法的重建图像进行了量化对比分析.结果显示，在相同的SMS加速倍数下，使用VIRGINIA方法进行重建的图像质量均好于RAKI方法，且远好于传统Slice-GRAPPA方法.

关键词： 磁共振图像重建; 多层同时成像; k空间插值鲁棒人工神经网络（RAKI）; 虚拟线圈; 卷积神经网络（CNN）

本文引用格式

王婉婷 , 苏适 , 贾森 , 梁栋 , 王海峰 . 基于虚拟线圈和卷积神经网络的多层同时激发图像重建[J]. 波谱学杂志, 2020 , 37(4) : 407 -421 . DOI: 10.11938/cjmr20202800

Abstract

This paper proposes an image reconstruction method for simultaneous multi-slice imaging (SMS) by combining the virtual conjugate coil (VCC) technology and robust artificial-neural-networks for k-space interpolation (RAKI). This method can effectively improve the reconstruction quality, and is named VIRGINIA (VIRtual conjuGate coIls Neural-networks InterpolAtion). VIRGINIA utilizes the complex conjugate symmetry property of the virtual coil concept to generate virtual coil data for training, and obtains better image quality by applying the trained network to the original aliased SMS data. With experimental data, the VIRGINIA method was compared to other reconstruction methods (i.e., RAKI only and slice-GRAPPA) in terms of quantitative indices such as structural similarity index (SSIM), peak signal-to-noise ratio (PSNR) and root mean square error (RMSE). The results demonstrated that, under some certain slice-acceleration factors, VIRGINIA produced better reconstruction quality than those obtainable by Slice-GRAPPA and RAKI.

Key words： magnetic resonance image reconstruction; simultaneous multi-slice imaging; robust artificial-neural-networks for k-space interpolation (RAKI); virtual conjugate coil; convolutional neural network (CNN)

参考文献

[1] ZHENG H R, WU Y, HE Q, et al. Fast and high-resolution magnetic resonance imaging on high field system[J]. Life Science Instruments, 2018, 16(Z1):29-44, 54.郑海荣, 吴垠, 贺强, 等. 基于高场磁共振的快速高分辨成像[J]. 生命科学仪器, 2018, 16(Z1):29-44, 54.
[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] 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.
[4] 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.
[5] SETSOMPOP K, GAGOSKI B A, POLIMENI J R, et al. Blipped-controlled aliasing in parallel imaging for simultaneous multislice echo planar imaging with reduced g-factor penalty[J]. Magn Reson Med. 2012, 67(5):1210-1224.
[6] LARKMAN D J, HAJNAL J V, HERLIHY A H, et al. Use of multicoil arrays for separation of signal from multiple slices simultaneously excited[J]. J Magn Reson Imaging, 2001, 13(2):313-317.
[7] BREUER F A, BLAIMER M, HEIDEMANN R M, et al. Controlled aliasing in parallel imaging results in higher acceleration (CAIPIRINHA) for multi-slice imaging[J]. Magn Reson Med, 2005, 53(3):684-691.
[8] NUNES R G, HAJNAL J V, GOLAY X, et al. Simultaneous slice excitation and reconstruction for single shot EPI[C]//14th Annual Meeting of ISMRM, Seattle, Washington, USA. 2006:293.
[9] CHENG J, WANG H, YING L, et al. Model learning:Primal dual networks for fast MR imaging[C]. International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI), Oct. 13-17, 2019, Shenzhen, China. Springer, Cham, 2019:21-29.
[10] WANG H, CHENG J, JIA S, et al. Accelerating MR imaging via deep Chambolle-Pock network[C]. 201941st Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), Jul. 23-27, 2019, Berlin, Germany. IEEE, 2019:6818-6821.
[11] CHENG H T, WANG S S, KE Z W, et al. A deep recursive cascaded convolutional network for parallel MRI[J]. Chinese J Magn Reson, 2019, 36(4):437-445.程慧涛, 王珊珊, 柯子文, 等. 基于深度递归级联卷积神经网络的并行磁共振成像方法[J]. 波谱学杂志, 2019, 36(4):437-445.
[12] WANG H Z, ZHAO D, YANG L Q, et al. An approach for training data enrichment and batch labeling in AI+MRI aided diagnosis[J]. Chinese J Magn Reson, 2018, 35(4):447-456.汪红志, 赵地, 杨丽琴, 等. 基于AI+MRI的影像诊断的样本增广与批量标注方法[J]. 波谱学杂志, 2018, 35(4):447-456.
[13] AKCAKAYA M, MOELLER S, WEINGAERTNER S, et al. Scan-specific robust artificial-neural-networks for k-space interpolation (RAKI) reconstruction:Database-free deep learning for fast imaging[J]. Magn Reson Med, 2019, 81(1):439-453.
[14] EO T, JUN Y, KIM T, et al. KIKI-net:cross-domain convolutional neural networks for reconstructing undersampled magnetic resonance images[J]. Magn Reson Med, 2018, 80(5):2188-2201.
[15] BLAIMER M, BREUER F A, SEIBERLICH N, et al. Accelerated volumetric MRI with a SENSE/GRAPPA combination[J]. J Magn Reson Imaging, 2006, 24(2):444-450.
[16] ZHANG C, MOELLER S, WEINGAERTNER S, et al. Accelerated simultaneous multi-slice MRI using subject-specific convolutional neural networks[C]. Conference Record of the Asilomar Conference on Signals Systems and Computers, Pacific Grove, CA. 2018:1636-1640.
[17] BLAIMER M, GUTBERLET M, KELLMAN P, et al. Virtual coil concept for improved parallel MRI employing conjugate symmetric signals[J]. Magn Reson Med, 2009, 61(1):93-102.
[18] KINGMA D P, BA J. Adam:A method for stochastic optimization[C]. International Conference on Learning Representations (ICLR), May 7-9, 2015, San Diego, USA. arXiv preprint arXiv:1412.6980, 2014.
[19] SU S, LU N, JIA L, et al. High spatial resolution BOLD fMRI using simultaneous multislice excitation with echo-shifting gradient echo at 7 Tesla[J]. Magnc Reson Imaging, 2020, 66:86-92.
[20] MICKEVICIUS N J, PAULSON E S, TUGAN MUFTULER L, et al. Application of a k-space interpolating artificial neural network to in-plane accelerated simultaneous multislice imaging[OL].[2019-02-01]. https://ui.adsabs.harvard.edu/abs/2019arXiv190208589M

Options

文章导航

模态框（Modal）标题

摘要

本文引用格式

Abstract

参考文献