2.845

2023影响因子

(CJCR)

  • 中文核心
  • EI
  • 中国科技核心
  • Scopus
  • CSCD
  • 英国科学文摘

留言板

尊敬的读者、作者、审稿人, 关于本刊的投稿、审稿、编辑和出版的任何问题, 您可以本页添加留言。我们将尽快给您答复。谢谢您的支持!

姓名
邮箱
手机号码
标题
留言内容
验证码

一种面向医学图像非刚性配准的多维特征度量方法

陆雪松 涂圣贤 张素

陆雪松, 涂圣贤, 张素. 一种面向医学图像非刚性配准的多维特征度量方法. 自动化学报, 2016, 42(9): 1413-1420. doi: 10.16383/j.aas.2016.c150608
引用本文: 陆雪松, 涂圣贤, 张素. 一种面向医学图像非刚性配准的多维特征度量方法. 自动化学报, 2016, 42(9): 1413-1420. doi: 10.16383/j.aas.2016.c150608
LU Xue-Song, TU Sheng-Xian, ZHANG Su. A Metric Method Using Multidimensional Features for Nonrigid Registration of Medical Images. ACTA AUTOMATICA SINICA, 2016, 42(9): 1413-1420. doi: 10.16383/j.aas.2016.c150608
Citation: LU Xue-Song, TU Sheng-Xian, ZHANG Su. A Metric Method Using Multidimensional Features for Nonrigid Registration of Medical Images. ACTA AUTOMATICA SINICA, 2016, 42(9): 1413-1420. doi: 10.16383/j.aas.2016.c150608

一种面向医学图像非刚性配准的多维特征度量方法

doi: 10.16383/j.aas.2016.c150608
基金项目: 

国家民委科研项目 14ZNZ024

国家自然科学基金 61002046

详细信息
    作者简介:

    涂圣贤 上海交通大学生物医学工程学院东方学者特聘教授.主要研究方向为心血管成像与定量分析.E-mail:sxtu@sjtu.edu.cn

    张素 上海交通大学生物医学工程学院副教授.主要研究方向为医学影像处理与分析.E-mail:suzhang@sjtu.edu.cn

    通讯作者:

    陆雪松 中南民族大学生物医学工程学院副教授.主要研究方向为医学影像配准、分割和辅助诊断.本文通信作者.E-mail:xslu-scuec@hotmail.com

A Metric Method Using Multidimensional Features for Nonrigid Registration of Medical Images

Funds: 

Scientific Research Projects by the State Ethnic Affairs Commission of China 14ZNZ024

National Natural Science Foundation of China 61002046

More Information
    Author Bio:

    Professor of special appointment at the School of Biomedical Engineering, Shanghai Jiao Tong University. His research interest covers cardiovascular imaging and quantitative analysis.E-mail:

    Associate professor at the School of Biomedical Engineering, Shanghai Jiao Tong University. Her research interest covers medical image processing and analysis.E-mail:

    Corresponding author: LU Xue-Song Associate professor at the College of Biomedical Engineering, South-Central University for Nationalities. His research interest covers medical image registration, segmentation, and assisted diagnosis. Corresponding author of this paper. E-mail:xslu-scuec@hotmail.com
  • 摘要: 医学图像的非刚性配准对于临床的精确诊疗具有重要意义.待配准图像对中目标的大形变和灰度分布呈各向异性给非刚性配准带来困难.本文针对这个问题,提出基于多维特征的联合Renyi α-entropy度量结合全局和局部特征的非刚性配准算法.首先,采用最小距离树构造联合Renyi α-entropy,建立多维特征度量新方法.然后,演绎出新度量准则相对于形变模型参数的梯度解析表达式,采用随机梯度下降法进行参数寻优.最终,将图像的Canny特征和梯度方向特征融入新度量中,实现全局和局部特征相结合的非刚性配准.通过在36对宫颈磁共振(Magnetic resonance,MR)图像上的实验,该方法的配准精度相比较于传统互信息法和互相关系数法有明显提高.这也表明,这种度量新方法能克服因图像局部灰度分布不一致造成的影响,一定程度地减少误匹配,为临床的精确诊疗提供科学依据.
  • 图  1  膀胱和宫颈放射治疗前后磁共振(Magnetic resonance, MR)图像对比

    Fig.  1  Contrast of magnetic resonance (MR) images before and after radiotherapy at the bladder and the cervix

    图  2  非刚性配准的框架图

    Fig.  2  The framework of nonrigid registration

    图  3  特征图像实例

    Fig.  3  Examples of the image features

    图  4  仿射粗配、互相关系数、互信息、基于图的广义互信息和本文方法的重叠率对比boxplot图(对于每一个解剖结构, 最左边的列显示了仿射粗配的结果, 第2列显示了互相关系数的结果, 第3列显示了互信息的结果, 第4列显示了基于图的广义互信息的结果, 最右边的列显示了本文方法的结果.一个星号表示两种方法重叠率中值在统计上的明显差异.)

    Fig.  4  The boxplot of overlap scores using affine matching, CC, MI, GMI and our proposal (For each anatomical structure, the leftmost column shows the result for affine matching. The second column shows the result for CC. The third column shows the result for MI. The fourth column shows the result for GMI. The rightmost column shows the result for our proposal. A star indicates a statistical significant difference of the median overlaps of the two methods.)

    图  5  配准结果示例, 参考图像与被形变的浮动图像采用Checkerboard模式融合

    Fig.  5  Demonstration of the registration result, and the reference image is combined with the deformed moving image, using a Checkerboard pattern

    表  1  所有配准DSC的均值和方差

    Table  1  The mean and variance of DSC about all registration results

    结构区域 配准方法 DSC值
    仿射粗配 0.7618±0.0123
    CC 0.8271±0.0056
    CTV MI 0.8332±0.0756
    GMI 0.8460±0.0706
    本文方法 0.8462±0.0684
    仿射粗配 0.6756±0.0198
    CC 0.7386±0.0200
    Bladder MI 0.7375±0.1362
    GMI 0.7697±0.1332
    本文方法 0.7800±0.1313
    仿射粗配 0.6861±0.0083
    CC 0.7338±0.0042
    Rectum MI 0.7410±0.0673
    GMI 0.7566±0.0627
    本文方法 0.7680±0.0577
    下载: 导出CSV
  • [1] Savage N. Technology: multiple exposure. Nature, 2013, 502(7473): S90-S91 doi: 10.1038/502S90a
    [2] Sotiras A, Davatzikos C, Paragios N. Deformable medical image registration: a survey. IEEE Transactions on Medical Imaging, 2013, 32(7): 1153-1190 doi: 10.1109/TMI.2013.2265603
    [3] Schmidt-Richberg A, Werner R, Handels H, Ehrhardt J. Estimation of slipping organ motion by registration with direction-dependent regularization. Medical Image Analysis, 2012, 16(1): 150-159 doi: 10.1016/j.media.2011.06.007
    [4] 张桂梅, 曹红洋, 储王君, 曾接贤.基于Nyström低阶近似和谱特征的图像非刚性配准.自动化学报, 2015, 41(2): 429-438 http://www.aas.net.cn/CN/abstract/abstract18621.shtml

    Zhang Gui-Mei, Cao Hong-Yang, Chu Jun, Zeng Jie-Xian. Non-rigid image registration based on low-rank Nyström approximation and spectral feature. Acta Automatica Sinica, 2015, 41(2): 429-438 http://www.aas.net.cn/CN/abstract/abstract18621.shtml
    [5] 周斌, 宋晓, 黄晓阳, 王博亮.联合直方图的刚性配准方法及其在肝脏CT图像配准中的应用.厦门大学学报(自然科学版), 2015, 54(3): 397-403 http://www.cnki.com.cn/Article/CJFDTOTAL-XDZK201503019.htm

    Zhou Bin, Song Xiao, Huang Xiao-Yang, Wang Bo-Liang.A rigid registration method based on joint distribution histogram and its application on liver CT images. Journal of Xiamen University (Natural Science), 2015, 54(3): 397-403 http://www.cnki.com.cn/Article/CJFDTOTAL-XDZK201503019.htm
    [6] 夏威, 高欣, 王雷, 周志勇, 张冉.基于点集与互信息的肺部CT图像三维弹性配准算法.江苏大学学报(自然科学版), 2014, 35(5):558-563 http://www.cnki.com.cn/Article/CJFDTOTAL-JSLG201405012.htm

    Xia Wei, Gao Xin, Wang Lei, Zhou Zhi-Yong, Zhang Ran. 3D deformable registration algorithm of CT lung images based on point set and mutual information. Journal of Jiangsu University (Natural Science Edition), 2014, 35(5): 558-563 http://www.cnki.com.cn/Article/CJFDTOTAL-JSLG201405012.htm
    [7] Cachier P, Bardinet E, Dormont D, Pennec X, Ayache N. Iconic feature based nonrigid registration: the PASHA algorithm. Computer Vision and Image Understanding, 2003, 89(2-3): 272-298 doi: 10.1016/S1077-3142(03)00002-X
    [8] Mellor M, Brady M. Phase mutual information as a similarity measure for registration. Medical Image Analysis, 2005, 9(4): 330-343 doi: 10.1016/j.media.2005.01.002
    [9] Loeckx D, Slagmolen P, Maes F, Vandermeulen D, Suetens P. Nonrigid image registration using conditional mutual information. IEEE Transactions on Medical Imaging, 2010, 29(1): 19-29 doi: 10.1109/TMI.2009.2021843
    [10] Neemuchwala H, Hero A, Carson P. Image matching using alpha-entropy measures and entropic graphs. Signal Processing, 2005, 85(2): 277-296 doi: 10.1016/j.sigpro.2004.10.002
    [11] Staring M, van der Heide U A, Klein S, Viergever M A, Pluim J P W. Registration of cervical MRI using multifeature mutual information. IEEE Transactions on Medical Imaging, 2009, 28(9): 1412-1421 doi: 10.1109/TMI.2009.2016560
    [12] Rivaz H, Karimaghaloo Z, Collins D L. Self-similarity weighted mutual information: a new nonrigid image registration metric. Medical Image Analysis, 2014, 18(2): 343-358 doi: 10.1016/j.media.2013.12.003
    [13] 冯兆美, 党军, 崔崤山尧, 焦阳.基于B样条自由形变三维医学图像非刚性配准研究.影像科学与光化学, 2014, 32(2): 200-208 http://www.cnki.com.cn/Article/CJFDTOTAL-GKGH201402009.htm

    Feng Zhao-Mei, Dang Jun, Cui Xiao-Yao, Jiao Yang. Bspline free-form deformation based 3-D non-rigid registration of medical images. Imaging Science and Photochemistry, 2014, 32(2): 200-208 http://www.cnki.com.cn/Article/CJFDTOTAL-GKGH201402009.htm
    [14] Lowe D G. Distinctive image features from scale-invariant keypoints. International Journal of Computer Vision, 2004, 60(2): 91-110 doi: 10.1023/B:VISI.0000029664.99615.94
    [15] Bay H, Tuytelaars T, van Gool L. SURF: speeded up robust features. In: Proceedings of the 9th European Conference on Computer Vision. Graz, Austria: Springer, 2006. 404-417
    [16] Tola E, Lepetit V, Fua P. DAISY: an efficient dense descriptor applied to wide-baseline stereo. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2010, 32(5):815-830 doi: 10.1109/TPAMI.2009.77
    [17] Canny J. A computational approach to edge detection. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1986, PAMI-8(6): 679-698 doi: 10.1109/TPAMI.1986.4767851
    [18] Neemuchwala H F, Hero A O. Image registration in high dimensional feature space. In: Proceedings of the 2005 SPIE 5674, Computational Imaging III. San Jose, CA: SPIE, 2005. 99-113
    [19] Sabuncu M R, Ramadge P J. Using spanning graphs for efficient image registration. IEEE Transactions on Image Processing, 2008, 17(5): 788-797 doi: 10.1109/TIP.2008.918951
    [20] Hero A O, Ma B, Michel O J J, Gorman J. Applications of entropic spanning graphs. IEEE Signal Processing Magazine, 2002, 19(5): 85-95 doi: 10.1109/MSP.2002.1028355
    [21] Klein S, Staring M, Pluim J P W. Evaluation of optimization methods for nonrigid medical image registration using mutual information and B-splines. IEEE Transactions on Image Processing, 2007, 16(12): 2879-2890 doi: 10.1109/TIP.2007.909412
    [22] Klein S, Staring M, Murphy K, Viergever M A, Pluim J P W. Elastix: a toolbox for intensity-based medical image registration. IEEE Transactions on Medical Imaging, 2010, 29(1): 196-205 doi: 10.1109/TMI.2009.2035616
    [23] Ibanez L, Schroeder W, Ng L, Cates J. The ITK software guide version 2.4 [Online], available: http://www.itk.org, September16, 2014
    [24] [24 Dice L R. Measures of the amount of ecologic association between species. Ecology, 1945, 26(3): 297-302 doi: 10.2307/1932409
    [25] Wilcoxon F. Individual comparisons by ranking methods.Biometrics Bulletin, 1945, 1(6): 80-83 doi: 10.2307/3001968
  • 加载中
图(5) / 表(1)
计量
  • 文章访问数:  2013
  • HTML全文浏览量:  477
  • PDF下载量:  1214
  • 被引次数: 0
出版历程
  • 收稿日期:  2015-10-08
  • 录用日期:  2016-03-10
  • 刊出日期:  2016-09-01

目录

    /

    返回文章
    返回