基于灰度的二维/三维图像配准方法及其在骨科导航手术中的实现  

Implementation of intensity-based 2D/3D image registration in orthopaedic navigation system

作　　者：张薇[1] 黄毓瑜[1] 栾胜[1] 张家磊[1] 

机构地区：[1]北京航空航天大学机械工程及自动化学院,北京100083

出　　处：《中国医学影像技术》2007年第7期1080-1084,共5页Chinese Journal of Medical Imaging Technology

基　　金：国家科技支撑计划(2006BAI03A16);国家杰出青年科学基金(60525314);北京市科技计划(H060720050230)资助

摘　　要：目的随着基于医学图像导航的手术对三维信息需求的加强,建立二维/三维的医学图像配准算法,在二维规划的基础上引入三维信息。方法首先,将获取的骨盆图像体数据集通过投影得到DRR图像,将获取的X线图像进行校正;然后,利用灰度相似性测度配准DRR图像和无失真X线图像。结果在初始位置相对坐标位移<20mm,角度<5°时配准精度小于1.47%,取得良好的配准结果。结论本文提出的基于灰度计算的配准技术能够实现术中二维医学图像与术前三维医学图像的配准,实现术中二维/三维手术规划相结合。Objective 2D/3D medical image X-ray image is got wit know the comparativel To import registration. h the correct y position of the 3 dimensional information into computer aided Orthopaedic surgery with the help of Methods Firstly, DRR image is got with the projection of CT dataset and the corrected ion method;Secondly, the intensity-based 2D/3D medical registration method is used to 2D and 3D images. Results Our method is effective when the initial transform （less than 20 ram） and rotation （less than 5 degree） are in certain range. Conclusion In this paper, we describe a computer aided navigation method, integrating image registration based on intensity-based method using pattern intensity similarity measurement, which can combine the 2D medical information with 3D information.

关 键 词：医学图像 DRR 图像配准 二维规划/三维规划 

分 类 号：R445[医药卫生—影像医学与核医学]