Pickering Mark R, Muhit Abdullah A, Scarvell Jennie M, Smith Paul N
School of Information Technology and Electrical Engineering, The University of New South Wales, Australian Defence Force Academy, Canberra, Australia.
Annu Int Conf IEEE Eng Med Biol Soc. 2009;2009:5821-4. doi: 10.1109/IEMBS.2009.5335172.
2D-3D image registration has been adopted in many clinical applications such as image-guided surgery and the kinematic analysis of bones in knee and ankle joints. In this paper we propose a new single-plane 2D-3D registration algorithm which requires far less iteration than previous techniques. The new algorithm includes a new multi-modal similarity measure and a novel technique for the analytic calculation of the required gradients. Our experimental results show that, when compared to existing gradient and non-gradient based techniques, the proposed algorithm has a wider range of initial poses for which registration can be achieved and requires significantly fewer iterations to converge to the true 3D position of the anatomical structure.
二维到三维图像配准已被应用于许多临床应用中，如图像引导手术以及膝关节和踝关节骨骼的运动分析。在本文中，我们提出了一种新的单平面二维到三维配准算法，该算法所需的迭代次数比以前的技术少得多。新算法包括一种新的多模态相似性度量和一种用于解析计算所需梯度的新技术。我们的实验结果表明，与现有的基于梯度和非梯度的技术相比，所提出的算法在可实现配准的初始姿态范围内更宽，并且收敛到解剖结构的真实三维位置所需的迭代次数明显更少。
