Fast Brain MRI Reconstruction Using Deep Iterative Convolutional Neural Network

doi:10.3778/j.issn.1673-9418.2004071

Abstract

Abstract:

Human brain MRI is usually multi-slice, and there is data redundancy between adjacent slices. Deep learning has become a powerful tool in the field of undersampled MRI reconstruction. However, the current reconstruction algorithms based on deep learning are mainly for a single MRI image. In order to make full use of the data redun-dancy in brain MRI data and obtain higher reconstruction quality and acceleration factor, a deep iterative convolu-tional neural network (DICNN) is proposed. In each iteration, a bi-directional convolution module (BDC) is used to explore the data redundancy between adjacent slices, and then a 2D convolution module (refine net, RNET) is used to further explore the data redundancy within a single MRI slice. Simulation experiments on a single-coil brain MRI dataset show that the proposed algorithm is better than the algorithm based on a single MRI image under different undersampling factors. This method can not only effectively make use of the data redundancy between brain MRI slices and recover more tissue structure details, but also meet real-time MRI reconstruction at a speed of 49 slices per second.

Key words: brain magnetic resonance imaging (MRI), deep learning, image reconstruction, convolutional neural network (CNN)

摘要：

人体脑部MRI通常是多切片的，并且相邻切片间存在数据冗余。深度学习已经成为欠采样MRI重建领域的有力工具，然而目前基于深度学习的重建算法主要是针对单幅MRI图像。为了充分利用脑部MRI数据中的数据冗余，以获取更高的重建质量与加速因子，提出了一种深度迭代卷积神经网络（DICNN）。在每次迭代中，首先使用双向卷积模块（BDC）探索相邻切片间的数据冗余，然后用2D卷积模块（RNET）进一步探索单幅MRI切片内部的数据冗余。在单线圈的脑部MRI数据集上的仿真实验表明，提出的重建算法在不同欠采样因子下的重建效果优于基于单幅MRI图像的重建算法。该方法不仅能够有效地利用脑部MRI切片间的数据冗余，恢复更多的组织结构细节，还能进行实时的MRI重建，速度可达每秒49张。

关键词: 脑部核磁共振成像（MRI）, 深度学习, 图像重建, 卷积神经网络（CNN）

DU Nianmao, SONG Wei. Fast Brain MRI Reconstruction Using Deep Iterative Convolutional Neural Network[J]. Journal of Frontiers of Computer Science and Technology, 2020, 14(12): 2150-2160.

杜年茂，宋威. 深度迭代卷积神经网络的快速脑部MRI重建算法[J]. 计算机科学与探索, 2020, 14(12): 2150-2160.

References

[1] Knoll F, Zbontar J, Sriram A, et al. fastMRI: a publicly avail-able raw k-space and DICOM dataset of knee images for accelerated MR image reconstruction using machine learning[J]. Radiology: Artificial Intelligence, 2020, 2(1): e190007.
[2] Yang G, Yu S, Dong H, et al. DAGAN: deep de-aliasing genera-tive adversarial networks for fast compressed sensing MRI reconstruction[J]. IEEE Transactions on Medical Imaging, 2017, 37(6): 1310-1321.
[3] Eo T, Jun Y, Kim T, et al. KIKI-net: cross-domain convolu-tional neural networks for reconstructing undersampled mag-netic resonance images[J]. Magnetic Resonance in Medicine, 2018, 80(5): 2188-2201.
[4] Donoho D L. Compressed sensing[J]. IEEE Transactions on Information Theory, 2006, 52(4): 1289-1306.
[5] Lustig M, Donoho D L, Santos J M, et al. Compressed sensing MRI[J]. IEEE Signal Processing Magazine, 2008, 25(2): 72-82.
[6] Chen B. A research of rapid magnetic resonance imaging based on compressed sensing[D]. Chengdu: University of Electronic Science and Technology of China, 2016.陈兵. 基于压缩感知的快速核磁成像算法研究[D]. 成都: 电子科技大学, 2016.
[7] Zhang J G. Fast MRI image reconstruction based on com-pressed sensing[D]. Harbin: Harbin University of Science and Technology, 2016.张建广. 基于压缩感知的快速MRI图像重建[D]. 哈尔滨：哈尔滨科技大学, 2016.
[8] Yan S Y. Fast algorithm for parallel magnetic resonance image reconstruction based on total variations model[D]. Nanjing: Nanjing University of Posts and Telecommunications, 2018.晏士友. 基于全变分的并行磁共振图像重建的快速算法研究[D]. 南京: 南京邮电大学, 2018.
[9] Li G Y, Hou X D, Zhou B J, et al. Reconstruction of com-pressed sensing MRI image based on discrete shearlet trans-form[J]. Application Research of Computer, 2013, 30(6):1895-1898.李国燕, 侯向丹, 周博君, 等. 基于离散剪切波的压缩感知MRI图像重建[J]. 计算机应用研究, 2013, 30(6): 1895-1898.
[10] Qu X, Guo D, Ning B, et al. Undersampled MRI recons-truction with patch-based directional wavelets[J]. Magnetic Resonance Imaging, 2012, 30(7): 964-977.
[11] Sun J, Li H, Xu Z. Deep ADMM-Net for compressive sensing MRI[C]//Proceedings of the 2016 Annual Conference on Neural Information Processing Systems, Barcelona, Dec 5-10, 2016. Red Hook: Curran Associates, 2016: 10-18.
[12] Song Y, Zhu Z, Lu Y, et al. Reconstruction of magnetic reso-nance imaging by three-dimensional dual-dictionary learning[J]. Magnetic Resonance in Medicine, 2014, 71(3): 1285-1298.
[13] Zhan Z, Cai J F, Guo D, et al. Fast multiclass dictionaries learning with geometrical directions in MRI reconstruction[J]. IEEE Transactions on Biomedical Engineering, 2015, 63(9): 1850-1861.
[14] Qin C, Schlemper J, Caballero J, et al. Convolutional recurrent neural networks for dynamic MR image reconstruction[J]. IEEE Transactions on Medical Imaging, 2018, 38(1): 280-290.
[15] Schlemper J, Caballero J, Hajnal J V, et al. A deep cascade of convolutional neural networks for MR image reconstruction[C]//LNCS 10265: Proceedings of the 25th International Con-ference on Information Processing in Medical Imaging, Boone, Jun 25-30, 2017. Berlin, Heidelberg: Springer, 2017: 647-658.
[16] Wang S, Cheng H, Ying L, et al. DeepcomplexMRI: exploiting deep residual network for fast parallel MR imaging with complex convolution[J]. Magnetic Resonance Imaging, 2020, 68: 136-147.
[17] Souza R, Lebel R M, Frayne R. A hybrid, dual domain, cascade of convolutional neural networks for magnetic reso-nance image reconstruction[C]//Proceedings of the 2019 International Conference on Medical Imaging with Deep Learning, London, Jul 8-10, 2019: 437-446.
[18] Mikolov T, Karafiát M, Burget L, et al. Recurrent neural network based language model[C]//Proceedings of the 11th Annual Conference of the International Speech Communi-cation Association, Makuhari, Sep 26-30, 2010: 1045-1048.
[19] Wang S, Yi L, Chen Q, et al. Edge-aware fully convolutional network with CRF-RNN layer for hip-pocampus segmentation[C]//Proceedings of the IEEE 8th Joint International Infor-mation Technology and Artificial Intelligence Conference, Chongqing, May 24-26, 2019. Piscataway: IEEE, 2019: 803-806.
[20] Xie K, Wen Y. LSTM-MA: a LSTM method with multi-moda-lity and adjacency constraint for brain image segmentation[C]//Proceedings of the 2019 IEEE International Conference on Image Processing, Taiwan, China, Sep 22-25, 2019. Piscat-away: IEEE, 2019: 240-244.
[21] Liu C, Xiao Z Y, Du N M. Application of improved convolu-tional neural network in medical image segmentation[J]. Journal of Frontiers of Computer Science and Technology,2019, 13(9): 1593-1603. 刘辰, 肖志勇, 杜年茂. 改进的卷积神经网络在医学图像分割上的应用[J]. 计算机科学与探索, 2019, 13(9): 1593-1603.
[22] Vaswani A, Shazeer N, Parmar N, et al. Attention is all you need[C]//Proceedings of the 2017 Annual Conference on Neural Information Processing Systems, Long Beach, Dec 4-9, 2017. Red Hook: Curran Associates, 2017: 5998-6008.
[23] Woo S, Park J, Lee J Y, et al. CBAM: convolutional block attention module[C]//LNCS 11211: Proceedings of the 15th European Conference on Computer Vision, Munich, Sep 8-14, 2018. Berlin, Heidelberg: Springer, 2018: 3-19.
[24] Wang X T, Chan K C K, Yu K, et al. EDVR: video restor-ation with enhanced deformable convolutional networks[C]//Proceedings of the 2019 IEEE Conference on Computer Vision and Pattern Recognition Workshops, Long Beach, Jun 16-20, 2019. Piscataway: IEEE, 2019: 1954-1963.
[25] Zhu X Z, Hu H, Lin S, et al. Deformable convnets v2: more deformable, better results[C]//Proceedings of the 2019 IEEE Conference on Computer Vision and Pattern Recognition, Long Beach, Jun 16-20, 2019. Piscataway: IEEE, 2019: 9308-9316.
[26] Zhang Y L, Li K P, Li K, et al. Image super-resolution using very deep residual channel attention networks[C]//LNCS 11211: Proceedings of the 15th European Conference on Computer Vision, Munich, Sep 8-14, 2018. Berlin, Heidel-berg: Springer, 2018: 294-301.
[27] Zu D L, G J H. Nuclear magnetic resonance imaging: physical principles and methods[M]. Beijing: Peking University Press, 2014.俎栋林, 高家红. 核磁共振成像: 物理原理和方法[M]. 北京: 北京大学出版社, 2014.
[28] Winkelmann S, Schaeffter T, Koehler T, et al. An optimal radial profile order based on the golden ratio for time-resolved MRI[J]. IEEE Transactions on Medical Imaging, 2007, 26(1): 68-76.
[29] Kim D, Adalsteinsson E, Spielman D M, et al. Simple analytic variable density spiral design[J]. Magnetic Resonance in Medicine, 2003, 50(1): 214-219.
[30] Bridson R. Fast Poisson disk sampling in arbitrary dimensions[C]//Proceedings of the 2007 International Conference on Computer Graphics and Interactive Techniques, San Diego, Aug 5-9, 2007. New York: ACM, 2007: 22.