Development of medical image registration technology using GPU

doi:10.11772/j.issn.1001-9081.2015.09.2486

Abstract

Abstract: The current medical image registration technology could not meet the real-time requirements for clinical diagnosis and treatment. Graphic Processing Unit (GPU) accelerated medical image registration technology was reviewed and discussed for this problem in this paper. The paper summarized GPU general purpose computation, studied current technology of medical image registration which based on GPU acceleration with the essential framework of medical image registration as main line, and implemented Positron Emission computed Tomography (PET) and Computed Tomography (CT) image registration experiments respectively on Central Processing Unit (CPU) and GPU computing platforms. The Normalized Mutual Information (NMI) value of GPU accelerated medical image registration based on Free Form Deformation ( FFD) and NMI was slightly smaller than that of CPU method, but the registration efficiency is 12 times than CPU method. Except keeping high registration accuracy, GPU accelerated medical image registration algorithms also get a lot of ascension in terms of registration speed.

Key words: medical imaging, image registration, Graphic Processing Unit (GPU), acceleration, spatial transform

摘要： 针对目前医学图像配准技术无法满足临床实时性需求问题,对基于图形处理器(GPU)加速的医学图像配准技术进行综述探讨。首先对GPU通用计算进行概述,再以医学图像配准基本框架为主线,对近年来基于GPU加速的医学图像配准技术在国内外发展现状进行深入研究,并针对正电子发射型计算机断层显像(PET)和电子计算机断层扫描(CT)数据的非线性配准问题,分别基于中央处理器(CPU)和GPU平台进行配准实验,通过实验结果的对比,体现GPU加速配准技术的优越性。基于GPU加速的自由形变(FFD)和归一化互信息(NMI)结合的非线性配准方法配准后互信息值略低于CPU平台的配准结果,但其配准速度是CPU平台的12倍。基于GPU加速的配准算法在保持配准精度的基础上,配准速度都得到了很大的提升。

关键词: 医学成像, 图像配准, 图形处理器, 加速, 空间变换

CLC Number:

TP391.41

ZHA Shanshan, WANG Yuanjun, NIE Shengdong. Development of medical image registration technology using GPU[J]. Journal of Computer Applications, 2015, 35(9): 2486-2491.

查珊珊, 王远军, 聂生东. 基于图形处理器加速的医学图像配准技术进展[J]. 计算机应用, 2015, 35(9): 2486-2491.

References

[1] HAJNAL J, HILL D, HAWKES D. Medical image registration [M]. New York: CRC Press, 2001:1.
[2] MARKELJ P, TOMAZEVIC D, LIKAR B, et al. A review of 3D/2D registration methods for image-guided interventions [J]. Medical Image Analysis, 2012,16(3):642-661.
[3] ROST R J, LICEAKANE B M, GINSBURG D, et al. OpenGL shading language [M]. London: Addison-Wesley Professional, 2009:1-34.
[4] LUNA F. Introduction to 3D game programming with DirectX 11 [M]. Boston: Mercury Learning and Information, 2012:15-17.
[5] EKLUND A, DUFORT P, FORSBERG D, et al. Medical image processing on the GPU-past, present and future [J]. Medical Image Analysis, 2013,17(8):1073-1094.
[6] PATTERSON D A, HENNESSY J L. Computer organization and design: the hardware/software interface [M]. San Francisco: Morgan Kaufmann Publishers, 2008:A13-A16.
[7] FLUCK O, VETTER C, WEIN W, et al. A survey of medical image registration on graphics hardware [J]. Computer Methods and Programs in Biomedicine, 2011,104(3):e45-e57.
[8] SHAMS R, SADEGHI P, KENNEDY R A, et al. A survey of medical image registration on multicore and the GPU [J]. IEEE Signal Processing Magazine, 2010,27(2):50-60.
[9] CRUM W, GRIFFIN L, HILL D, et al. Zen and the art of medical image registration: correspondence, homology, and quality [J]. NeuroImage, 2003,20(3):1425-1437.
[10] LOURENCO L H, WEINGAERTNER D, TODT E. Efficient implementation of Canny edge detection filter for ITK using CUDA [C]//Proceedings of the 2012 13th Symposium on Computing Systems. Piscataway: IEEE, 2012:33-40.
[11] QIN A, XU J, FENG Q, et al. Fast 3D rigid medical image registration based on GPU [J]. Application Research of Computers, 2010, 27(3): 1198-1200. (秦安, 徐建, 冯前进,等. 基于GPU的快速三维医学图像刚性配准技术[J]. 计算机应用研究, 2010, 27(3): 1198-1200.)
[12] GREVE D N, FISCHL B. Accurate and robust brain image alignment using boundary-based registration [J]. NeuroImage, 2009,48(1):63-72.
[13] WANG B, ZHU Z, MENG L. CUDA-based acceleration of three dimensional 3D medical images registration [J]. Journal of Chinese Computer Systems, 2013,34(11):2621-2625.(王蓓蕾,朱志良,孟琭.基于CUDA加速的三维医学图像配准[J].小型微型计算机系统,2013,34(11):2621-2625.)
[14] BANERJEE J, KLINK C, PETERS E D, et al. Fast and robust 3D ultrasound registration-block and game theoretic matching [J]. Medical Image Analysis, 2015,20(1):173-183.
[15] DEMIRCI S, BAUST M, KUTTER O, et al. Disocclusion-based 2D-3D registration for aortic interventions [J]. Computers in Biology and Medicine, 2013,43(4):312-322.
[16] SPOERK J, GENDRIN C, WEBER C, et al. High-performance GPU-based rendering for real-time, rigid 2D/3D-image registration and motion prediction in radiation oncology [J]. Zeitschrift Fur Medizinische Physik, 2012,22(1):13-20.
[17] MEMBARTH R, HANNIG F, TEICH J, et al. Frameworks for GPU accelerators: a comprehensive evaluation using 2D/3D image registration [C]//Proceedings of the 2011 IEEE 9th Symposium on Application Specific Processors. Piscataway: IEEE, 2011:78-81.
[18] GALLEA R, ARDIZZONE E, PIRRONE R, et al. Three-dimensional fuzzy kernel regression framework for registration of medical volume data [J]. Pattern Recognition, 2013,46(11):3000-3016.
[19] ZHANG Y, ZHOU P, REN Y, et al. GPU-accelerated large-size VHR images registration via coarse-to-fine matching [J]. Computers and Geosciences, 2014,66(5):54-65.
[20] HUANG Y, LIU J, TU M, et al. Research on CUDA-based SIFT registration of SAR image [C]//Proceedings of the 2011 4th International Symposium on Parallel Architectures, Algorithms and Programming. Piscataway: IEEE, 2011:100-104.
[21] RUECKERT D, SONODA L I, HAYES C, et al. Nonrigid registration using free-form deformations: application to breast MR images [J]. IEEE Transactions on Medical Imaging, 1999,18(8):712-721.
[22] MODAT M, RIDGWAY G R, TAYLOR Z A, et al. Fast free-form deformation using graphics processing units [J]. Computer Methods and Programs in Biomedicine, 2010,98(3):278-284.
[23] ANSORGE R E, SAWIAK S J, WILLIAMS G B. Exceptionally fast non-linear 3D image registration using GPUs [C]//Proceedings of the 2009 IEEE Nuclear Science Symposium Conference Record. Piscataway: IEEE, 2009:3207-3213.
[24] GRUSLYS A, SAWIAK S, ANSORGE R. 3000 non-rigid medical image registrations overnight on a single PC [C]//Proceedings of the 2011 IEEE Nuclear Science Symposium and Medical Imaging Conference. Piscataway: IEEE, 2011:3073-3080.
[25] THIRION J P. Image matching as a diffusion process: an analogy with Maxwell's demons [J]. Medical Image Analysis, 1998,2(3):243-260.
[26] SHARP G C, KANDASAMY N, FOLKERT H, et al. GPU-based streaming architectures for fast cone-beam CT image reconstruction and demons deformable registration [J]. Physics in Medicine and Biology, 2007,52(19):5771-5783.
[27] COURTY N, HELLIER P. Accelerating 3D non-rigid registration using graphics hardware [J]. International Journal of Image and Graphics, 2008,8(1):1-18.
[28] MUYAN-OZCELIK P, OWENS J D, XIA J, et al. Fast deformable registration on the GPU: a CUDA implementation of demons [C]//Proceedings of the 2008 International Conference on Computational Sciences and Its Applications. Piscataway: IEEE, 2008:223-233.
[29] NOE K Ø, SENNEVILLE B D, ELSTRØM U V, et al. Acceleration and validation of optical flow based deformable registration for image-guided radiotherapy [J]. Acta Oncologica, 2008,47(7):1286-1293.
[30] HUANG Y, TONG T, LIU W, et al. Accelerated diffeomorphic non-rigid image registration with CUDA based on demons algorithm [C]//Proceedings of the 2010 4th International Conference on Bioinformatics and Biomedical Engineering. Piscataway: IEEE, 2010:1-4.
[31] JOLDES G R, WITTEK A, MILLER K. Real-time nonlinear finite element computations on GPU-application to neurosurgical simulation [J]. Computer Methods in Applied Mechanics and Engineering, 2010,199(47/48):3305-3314.
[32] MOUSAZADEH H, MARAMI B, PATRICIU S, et al. GPU implementation of a deformable 3D image registration algorithm [C]//Proceedings of the 2011 Annual International Conference of the IEEE Engineering in Medicine and Biology Society. Piscataway: IEEE, 2011:4897-4900.
[33] CAI Y, GUO X, ZHONG Z, et al. Dynamic meshing for deformable image registration [J]. Computer-Aided Design, 2015,58(1):141-150.
[34] KNUTSSON H, ANDERSSON M. Morphons: segmentation using elastic canvas and paint on priors [C]//Proceedings of the 2005 International Conference on Image Processing. Piscataway: IEEE, 2005:1226-1229.
[35] WRANGSJO A, PETTERSSON J, KNUTSSON H. Non-rigid registration using Morphons [C]//Proceedings of the 14th Scandinavian Conference. Piscataway: IEEE, 2005:501-510.
[36] EKLUND A, ANDERSSON M, KNUTSSON H. Phase-based volume registration using CUDA [C]//Proceedings of the 2010 IEEE International Conference on Acoustics Speech and Signal Processing. Piscataway: IEEE, 2010:658-661.
[37] FORSBERG D, EKLUND A, ANDERSSON M, et al. Phase-based non-rigid 3D image registration: from minutes to seconds using CUDA [EB/OL]. [2015-01-03]. http://www.researchgate.net/publication/266057340_Phase-Based_Non-Rigid_3D_Image_Registration_From_Minutes_to_Seconds_Using_CUDA.
[38] STUDHOLME C, HILL D, HAWKES D J. An overlap invariant entropy measure of 3D medical image alignment [J]. Pattern Recognition, 1999, 32(1): 71-86.
[39] SHAMS R, BARNES N. Speeding up mutual information computation using NVIDIA CUDA hardware [C]//Proceedings of the 9th Biennial Conference of the Australian Pattern Recognition Society on Digital Image Computing Techniques and Applications. Washington, DC: IEEE Computer Society, 2007:555-560.
[40] CHEN S, QIN J, XIE Y, et al. CUDA-based acceleration and algorithm refinement for volume image registration [C]//Proceedings of the 2009 International Conference on Future BioMedical Information Engineering. Piscataway: IEEE, 2009:544-547.
[41] SHAMS R, SADEGHI P, KENNEDY R, et al. Parallel computation of mutual information on the GPU with application to real-time registration of 3D medical images [J]. Computer Methods and Program in Biomedicine, 2010,99(2):133-146.
[42] CHENG W, LU C. Acceleration of medical image registration using graphics process units in computing normalized mutual information [C]//Proceedings of the 2009 5th International Conference on Image and Graphics. Piscataway: IEEE, 2009:814-818.
[43] YANG J, HAN F, FEN C, et al. An accelerative method for multimodality medical image registration based on CUDA [C]//Proceedings of the 2011 4th International Congress on Image and Signal Processing. Piscataway: IEEE, 2011:1817-1821.
[44] IKEDA K, INO F, HAGIHARA K. Efficient acceleration of mutual information computation for nonrigid registration using CUDA [J]. IEEE Journal of Biomedical and Health Informatics, 2014,18(3):956-968.
[45] FUERST B, WEIN W, MVLLER M, et al. Automatic ultrasound-MRI registration for neurosurgery using the 2D and 3D LC² metric [J]. Medical Image Analysis, 2014,18(8):1312-1319.
[46] RIVEST-HENAULT D, DOWSON N, GREER B P, et al. Robust inverse-consistent affine CT-MR registration in MRI-assisted and MRI-alone prostate radiation therapy [J]. Medical Image Analysis, 2015,23(1):56-69.
[47] JALLOUL M, BAYDOUN M, AL-ALAOUI M A. Gauss-Newton image registration with CUDA [C]//Proceedings of the 2011 18th IEEE International Conference on Electronics, Circuits and Systems. Piscataway: IEEE, 2011:305-309.
[48] MICHALEK J, CAPEK M, JANACE J, et al. Matching of irreversibly deformed images in microscopy based on piecewise monotone subgradient optimization using parallel processing [C]//Proceedings of the 2012 IEEE Nuclear Science Symposium and Medical Imaging Conference. Piscataway: IEEE, 2012:3956-3963.
[49] FEI Y, RONG G D, WANG B, et al. Parallel L-BFGS-B algorithm on GPU [J]. Computers and Graphics, 2014,40:1-9.