基于“9点3面”配准方案的CT+MR异机三维图像融合研究

中国临床解剖学杂志 ›› 2011, Vol. 29 ›› Issue (4): 418-422.

基于“9点3面”配准方案的CT+MR异机三维图像融合研究

彭鳒侨¹，　鞠向阳²，　白　波¹，　刘　琦¹，　陈　艺¹，　莫建文¹，　朱巧洪³，　李新春³

广州医学院第一附属医院 1.关节外科广东省矫形植入骨科重点实验室, 3.放射科; 广州 510120;
2.英国Glasgow大学生物物理工程部，G23JZ U.K

收稿日期:2011-12-31 发布日期:2011-07-26
通讯作者: 白　波，教授，博士生导师，Tel: （020）83062680，E-mail: BaiB@gyfyy.com E-mail:pengjq@gyfyy.com
作者简介:彭鳒侨(1967-),男,广东人,博士生,高级工程师,研究方向：数码骨科
基金资助:
广州医学院2009年度留学回国人员、博士专项基金

3D-image fusion research of cross CT+MR modality based on localization registration approach of"9-point & 3-plane"

PENG Jian-qiao Matthew¹，JU Xiang-yang²， BAI Bo¹， LIU Qi¹， CHEN Yi¹， MO Jian-wen¹， ZHU Qiao-hong³， LI Xin-chun³

1. Department of Joint Surgery, The First Affiliated Hospital of Guangzhou Medical College, Orthopedic implantation key lab of Guangdong Province, Guangzhou 510120, China; 2. Department of Clinical Physics and Bio-Engineering, University of Glasgow, Glasgow G23JZ U.K; 3. Department of Radiology, The First Affiliated Hospital of Guangzhou Medical College, Guangzhou 510120, China

Received:2011-12-31 Published:2011-07-26

摘要/Abstract

摘要：

目的　尝试一种基于体表定位的二维图像配准方法,实现CT和MR异机三维图像的精确融合。方法　输入CT/MR原始数据后采用数字化格式转换,设计“9点3面”立体对位法进行配准,在实时工作站Mimics按照信息交互自动融合模式,通过讯号叠加技术完成图像融合。结果　以患者的头、膝为实例试验[CT+MR]立体图像的异机融合,生成了同时展现头部软硬组织、膝部病变性质和位置的互补影像，携带着来自CT和MR各自的讯号特征和医学信息，既能了解MR所发现的异常组织的明确位置，又能鉴别CT所发现的异常病灶的性质。结论　这种异机融合手段是对目前这一空缺技术的补充，同时，这一实验也将为进一步研制[CT+MR]同机三维融合设备提供经验借鉴。

关键词: 正电成像/核磁共振/计算机断层造影术, 三维重建, 定位配准, 异机图像融合

Abstract:

Objective　To attempt a localization registration approach of 2-Dimension (2D) images based on somatotopic localization to realize accurate fusion from cross modality of 3-Dimension (3D) CT & MR images.　Methods　To fix digital format after original data of CT/MR input, design cubic localization solution of"9-point & 3-plane"for registration, complete fusion at real-time workstation Mimics based on auto-fusing style of information exchanged by signal overlaid technique.　Results　The fused cubic images of cross [CT+MR] modality were mutually practiced by patients' cranium and knee samples, while complementary images of distinguishing soft and hard tissue in cranium and knee were created, which carrying signal characteristic and medical information respectively from CT & MR individually, and were helpful not only to learn specific location of abnormal organs found out by MR, but also to identify focal nature of abnormal lesions found out by CT.　Conclusions　This cross modality fusion scheme is a supplement for the technique vacant at present, meanwhile, this experiment will also provide experience drawn on invention of [CT+MR] single modality equipment of 3D imaging in advance.

Key words: PT/MR/CT, 3D rebuilt, Localization registration, Cross modality image fusion

中图分类号:

R814

彭鳒侨, 鞠向阳, 白　波, 刘　琦, 陈　艺, 莫建文, 朱巧洪, 李新春. 基于“9点3面”配准方案的CT+MR异机三维图像融合研究[J]. 中国临床解剖学杂志, 2011, 29(4): 418-422.

BANG Jian-Jiao, JU Xiang-Yang, BAI 　Bei, LIU 　Qi, CHEN 　Yi, MO Jian-Wen, SHU Qiao-Hong, LI Xin-Chun. 3D-image fusion research of cross CT+MR modality based on localization registration approach of"9-point & 3-plane"[J]. Chinese Journal Of Clinical Anatomy, 2011, 29(4): 418-422.

参考文献

[1] 张宏军, 范顺武, 方向前,等.腰椎退变与不稳的影像学研究
[J]．中国临床解剖学杂志,2006，24（3）：279-282.

[2] 余霞，葛红，李彬，等．医学图像融合的并行实现
[J]．计算机工程与应用,2008，44（34）：172-174.

[3] 林有籁，刘月华，王冬梅，等.CT 和MRI 图像融合三维重建颞下颌关节的研究
[J].华西口腔医学杂志,2008,26(2), 140-143.

[4] 岳晋，杨汝良，宦若虹，等. 基于双密度双树复数小波变换的图像融合研究
[J]．电子与信息学报,2009，31（1）：130-132.

[5] 杨立才，刘延梅，刘欣, 等.基于小波包变换的医学图像融合方法
[J]．中国生物医学工程学报,2009，28（1）：11-13.

[6] 罗火灵，许永忠，冉洋, 等．CT与 MR 医学图像的三种融合方法对比研究
[J]．中国    医疗器械信息,2008（10）：56-59.

[7] 李新春，方学文，陈键宇，等.兔坐骨神经挤压伤的MRI与SEP对比研究
[J].中国中西医结合影像杂志，2007，5（3）：178-180.

[8] 张宏军, 范顺武, 方向前, 等.腰椎退变与不稳的影像学研究
[J].中国临床解剖学杂志, 2006，24（3）：279-282.

[9] 彭鳒侨. 智能剪刀系列之1: 虚拟手术训练工程设计.;智能剪刀系列之2:脊椎手术仿真工程设计.中国版权局著作产权号码 2009-A-022593

[10]Ashburner J. A fast diffeomorphic image registration algorithm
[J]. Neuroimage,2007,38(1):95-113.

[11]Reyes NH, Messom CH. (2005) Identifying colour objectswith fuzzy colour contrast fusion
[C].3rd International conference on computational intelligence, robotics and autonomous systems, and FIRA roboworld congress.

[12] Camara O, Delso G, Colliot O, et al. Explicit incorporation of prior anatomical information into a nonrigid registration of thoracic and abdominal CT and 18-FDG whole-body emission PETimages
[J]. IEEE Trans Med Imaging, 2007, 26(2):164-178.

[13] Xiangyang JU, Siebert JP, Khambay BS，et al. Self-correction of 3D reconstruction from multi-view stereo images
[C]. The 2009 IEEE International Conference. Kyoto, Japan, October 3-4, 2009:1606-1613.

[14] Xiang G, Xiangyang JU, Holt P，et al. An automated approach to establishing dense correspondence for 3D facial models. Symposium on facial analysis and animation
[C]. BMVA, Edinburgh, June 2009.

[15] Xiang G, Xiangyang JU, Holt P, et al. Facial expression transferring with a    deformable model
[C]. Theory and Practice of Computer Graphics TPCG 2009, Cardiff University, June 17-19, 2009:117-124.