混合样条大尺度医学图像弹性配准算法研究
|
摘要
针对大尺度形变医学图像配准速度慢和精度低的特点,提出一种结合薄板样条(TPS)和B样条的弹性配准方法。该方法采用尺度不变特征变换算法(SIFT)进行图像特征提取与匹配,利用TPS算法将特征点对作为输入进行预处理,以降低浮动图像的形变尺度,从而提高下一步B样条配准的速度与精度。然后使用局部区域细化层次B样条方法将TPS生成的较稀疏的形变网格作为初始网格,结合有限记忆优化算法(L-BFGS)对控制网格作进一步处理,此过程只对形变较大的局部区域进行细化,以实现与参考图像的快速精确配准。实验结果表明,该方法较层次B样条方法有效地提高了配准的速度和精度。
In order to improve large scale deformation medical image registration efficiency and accuracy,this paper proposed a new elastic registration method based on thin plate spline( TPS) and B-spline. Firstly,the method used scale invariant feature transform( SIFT) algorithm to obtain the matched feature points. Secondly,it used TPS to reduce the scale of deformation. Thirdly,it used local region process hierarchical B-spline to treat the deformation grid as the initial grid,and it adopted the limi-ted memory Broyden Fietcher Goldfarb Shanno( L-BFGS) algorithm to complete registration quickly and accurately.This process only refined the local region which had a large deformation. Experimental results demonstrate that the efficiency and accuracy of the proposed method are superior to those of hierarchical B-spline.
In order to improve large scale deformation medical image registration efficiency and accuracy,this paper proposed a new elastic registration method based on thin plate spline( TPS) and B-spline. Firstly,the method used scale invariant feature transform( SIFT) algorithm to obtain the matched feature points. Secondly,it used TPS to reduce the scale of deformation. Thirdly,it used local region process hierarchical B-spline to treat the deformation grid as the initial grid,and it adopted the limi-ted memory Broyden Fietcher Goldfarb Shanno( L-BFGS) algorithm to complete registration quickly and accurately.This process only refined the local region which had a large deformation. Experimental results demonstrate that the efficiency and accuracy of the proposed method are superior to those of hierarchical B-spline.
引文
[1]Sotiras A,Davatzikos C,Paragios N.Deformable medical image registration:a survey[J].IEEE Trans on Medical Imaging,2013,32(7):1153-1190.
[2]Markelj P,Tomaevic 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]赵增军,李宝生.基于模型的医学图像形变配准在图像引导放疗中的应用研究进展[J].中国图象图形学报,2009,14(8):1504-1509.
[4]Pluim J P W,Maintz J B A,Viergever M A.Mutual-informationbased registration of medical images:a survey[J].IEEE Trans on Medical Imaging,2003,22(8):986-1004.
[5]Bookstein F L.Principal warps:thin-plate splines and the decomposition of deformations[J].IEEE Trans on Pattern Analysis and Machine Intelligence,1989,11(6):567-585.
[6]Rohr K,Stiehl H S,Sprengel R,et al.Landmark-based elastic registration using approximating thin-plate splines[J].IEEE Trans on Medical Imaging,2001,20(6):526-534.
[7]Chun S Y,Fessler J A.A simple regularizer for B-spline nonrigid image registration that encourages local invertibility[J].IEEE Journal of Selected Topics in Signal Processing,2009,3(1):159-169.
[8]Jacobson T J,Murphy M J.Optimized knot placement for B-splines in deformable image registration[J].Joint AAPMCOMP Meeting Program,2011,38(6):334-338.
[9]庞皓文,孙小杨,杨波,等.基于B样条的弹性配准在放疗CT图像间匹配的应用[J].中国医学物理学杂志,2011,28(5):2887-2889.
[10]申艳平.医学图像配准技术[J].中国医学物理学杂志,2013,30(1):3885-3889.
[11]Zhang Hongying,Zhang Jiawan,Sun Jizhou,et al.Non-rigid image registration algorithm based on B-splines approximation[J].Transaction of Tianjin University,2007,13(6):447-451.
[12]Zhuang Xiahai,Arridge S,Hawkes D J,et al.A nonrigid registration framework using spatially encoded mutual information and free-form deformations[J].IEEE Trans on Medical Imaging,2011,30(10):1819-1828.
[13]李彬,欧陕兴,田联房,等.基于自适应自由变形法和梯度下降法的胸部多模医学图像配准[J].计算机应用研究,2009,26(10):3978-3982.
[14]Huo Ju,Yang Ning,Cao Maoyong,et al.A reliable algorithm for image matching based on SIFT[J].Journal of Harbin Institute of Technology,2012,19(4):90-95.
[15]Goncalves H,Corte-Real L,Goncalves J A.Automatic image registration through image segmentation and SIFT[J].IEEE Trans on Geoscience and Remote Sensing,2011,49(7):2589-2600.
[16]Morales J L.A numerical study of limited memory BFGS method[J].Journal of Applied Mathematics,2002,15(4):481-487.
[2]Markelj P,Tomaevic 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]赵增军,李宝生.基于模型的医学图像形变配准在图像引导放疗中的应用研究进展[J].中国图象图形学报,2009,14(8):1504-1509.
[4]Pluim J P W,Maintz J B A,Viergever M A.Mutual-informationbased registration of medical images:a survey[J].IEEE Trans on Medical Imaging,2003,22(8):986-1004.
[5]Bookstein F L.Principal warps:thin-plate splines and the decomposition of deformations[J].IEEE Trans on Pattern Analysis and Machine Intelligence,1989,11(6):567-585.
[6]Rohr K,Stiehl H S,Sprengel R,et al.Landmark-based elastic registration using approximating thin-plate splines[J].IEEE Trans on Medical Imaging,2001,20(6):526-534.
[7]Chun S Y,Fessler J A.A simple regularizer for B-spline nonrigid image registration that encourages local invertibility[J].IEEE Journal of Selected Topics in Signal Processing,2009,3(1):159-169.
[8]Jacobson T J,Murphy M J.Optimized knot placement for B-splines in deformable image registration[J].Joint AAPMCOMP Meeting Program,2011,38(6):334-338.
[9]庞皓文,孙小杨,杨波,等.基于B样条的弹性配准在放疗CT图像间匹配的应用[J].中国医学物理学杂志,2011,28(5):2887-2889.
[10]申艳平.医学图像配准技术[J].中国医学物理学杂志,2013,30(1):3885-3889.
[11]Zhang Hongying,Zhang Jiawan,Sun Jizhou,et al.Non-rigid image registration algorithm based on B-splines approximation[J].Transaction of Tianjin University,2007,13(6):447-451.
[12]Zhuang Xiahai,Arridge S,Hawkes D J,et al.A nonrigid registration framework using spatially encoded mutual information and free-form deformations[J].IEEE Trans on Medical Imaging,2011,30(10):1819-1828.
[13]李彬,欧陕兴,田联房,等.基于自适应自由变形法和梯度下降法的胸部多模医学图像配准[J].计算机应用研究,2009,26(10):3978-3982.
[14]Huo Ju,Yang Ning,Cao Maoyong,et al.A reliable algorithm for image matching based on SIFT[J].Journal of Harbin Institute of Technology,2012,19(4):90-95.
[15]Goncalves H,Corte-Real L,Goncalves J A.Automatic image registration through image segmentation and SIFT[J].IEEE Trans on Geoscience and Remote Sensing,2011,49(7):2589-2600.
[16]Morales J L.A numerical study of limited memory BFGS method[J].Journal of Applied Mathematics,2002,15(4):481-487.