• 中国计算机学会会刊
  • 中国科技核心期刊
  • 中文核心期刊

计算机工程与科学

• 图形与图像 • 上一篇    下一篇

联合弯曲能量和标志点对应约束的非刚性医学图像配准

朱天宇,全惠敏,刘国才   

  1. (湖南大学电气与信息工程学院,湖南 长沙 410082)
  • 收稿日期:2018-06-25 修回日期:2018-09-26 出版日期:2019-05-25 发布日期:2019-05-25
  • 基金资助:

    国家自然科学基金(61671204);湖南省科技计划重点研发专项(2016WK2001)

Non-rigid medical image registration with bending
 energy and corresponding constraints of landmarks

ZHU Tianyu,QUAN Huimin,LIU Guocai   

  1. (College of Electrical and Information Engineering,Hunan University,Changsha 410082,China)
  • Received:2018-06-25 Revised:2018-09-26 Online:2019-05-25 Published:2019-05-25

摘要:

由于缺乏图像几何空间约束,基于互信息的非刚性医学图像配准常常产生不合理的形变。提出一种联合弯曲能量和标志点对应约束的非刚性医学图像配准方法,在互信息配准目标函数中添加弯曲能量惩罚和对应标志点间欧氏距离2个正则项,约束医学图像软组织不合理形变。脑部MRI、头颈部CT、胸部CBCT影像配准实验结果表明,该方法可有效提高配准质量。

关键词: 互信息, 弯曲能量, 标志点约束, 非刚性配准, 医学图像配准

Abstract:

Non-rigid medical image registration based on mutual information often leads to unreasonable deformation due to the lack of geometric space constraints in images. We therefore propose a non-rigid medical image registration method combining bending energy and corresponding constraints of landmarks. In order to restrain the unreasonable deformation of soft tissues in medical images, two regular terms including bending energy penalty and Euclidean distance between corresponding landmarks are added in the mutual information registration target function. The image registration results of brain MRI, head and neck CT and chest CBCT show that this method can improve the quality of registration effectively.
 

Key words: mutual information, bending energy, constraint of landmark, non-rigid registration, medical image registration