摘 要：本文提出了一种基于GTM（Graph Transformation Matching）和多分辨率B 样条的医学图像配准方法。为了得到精确的特征点对，使用GTM 算法去除血管树模型中的冗余点；然后使用多分辨率B 样条函数对图像进行配准。肝脏MRI 上的实验证明，本文提出的非刚性配准方法能够有效提取出特征点，并且达到了较好的配准精度。
关键词：医学图像配准；GTM；多分辨率B 样条
中图分类号：TP391.41 文献标识码：A 文章编号：2096-4706（2019）15-0085-04
Medical Image Registration Method Based on GTM and Multi-resolution B-splines
YU Jia1，ZHANG Xingwei2
（1.School of Computer and Information Engineering，Luoyang Institute of Science and Technology，Luoyang 471023，China；2.Luoyang Laipson Information Technology Co.，Ltd.，Luoyang 471003，China）
Abstract：This paper proposes a Graph Transformation Matching （GTM） and multilevel B-splines based approach for medical image registration. We utilize GTM algorithm to remove outliers to get the accurate pair-wise feature points. Then Multilevel B-Splines algorithm is used to find the non-linear transformation between two images. Experiments on liver MRI show that the proposed non-rigid registration method can effectively extract feature points and achieve better registration accuracy.
Keywords：medical image registration；GTM；multilevel B-splines
基金项目：本文系国家自然科学基金项目：协作式移动群体感知数据选择方法研究（项目编号：61602230）；洛阳理工学院校内基金项目：非刚性医学图像配准方法研究（项目编号：21010270）资助。
参考文献：
[1] Can A，Stewart C V，Roysam B，et al. A feature-based，robust，hierarchical algorithm for registering pairs of images of the curved human retina [J].IEEE Transactions on Pattern Analysis and Machine Intelligence，2002，24（3）：347-364.
[2] 冯林，张名举，贺明峰，等. 用分层互信息和薄板样条实现医学图像弹性自动配准 [J]. 计算机辅助设计与图形学学报，2005（7）：1492-1496.
[3] Rohr K，Stiehl S H，Sprengel R，et al. Landmark-Based Elastic Registration Using Approximating Thin-Plate Splines [J].IEEETrans. Med. Imaging，2001，20（6）：526-534.
[4] Rueckert D，Sonoda L I，Hayes C，et al. Nonrigid RegistrationUsing Free-Form Deformations：Application to Breast MR Images [J].IEEE Transactions on Medical Imaging，1999，18（8）：712-721.
[5] Niu C. Medical Image Registration Based on Mutual Information Using Kriging Probability Density Estimation [J].Conference proceedings：Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Conference，2005，3：3097-3099.
[6] Liu J，Tian J，Dai Y. Multi-modal medical image registration based on adaptive combination of intensity and gradient field mutual information [J].Conference proceedings ：Annual International Conference of the IEEE Engineering in Medicine and Biology Society.IEEE Engineering in Medicine and Biology Society. Conference，2006：1429-1432.
[7] Christensen G E，Rabbitt R D，Miller M I. Deformable templates using large deformation kinematics [J].IEEE Transactions on Image Processing，1996，5（10）：1435-1447.
[8] Christensen G E，Miller M I，Vannier M W，et al.Individualizing neuro-anatomical atlases using a massively parallel computer [J].Computer，1996，29（1）：32-38.
[9] Lester H，Arridge S R，Jansons K M，et al. Non-linear Registration with the Variable Viscosity Fluid Algorithm [A].Proceedings of International Conference on Information Processing in Medical Imaging [C].Berlin：Springer，1999：238-251.
[10] Hellier P，Barillot C. Cooperation between local and global approaches to register brain images [A].Proceedings of International Conference on Information Processing in Medical Imaging [C].Berlin：Springer，2001：315-328.
作者简介：
于佳（1980.10-），男，汉族，河南洛阳人，讲师，博士研究生，研究方向：医学图像处理、自然语言处理、机器学习；
张兴伟（1981.06-），男，汉族，河南洛阳人，研究员，硕士研究生，研究方向：医学图像处理、大数据技术。