Beg Mirza Faisal, Khan Ali
Medical Image Analysis Laboratory, School of Engineering Science, Simon Fraser University, Burnaby, BC V5A 1S6, Canada.
IEEE Trans Med Imaging. 2007 Sep;26(9):1179-89. doi: 10.1109/TMI.2007.898813.
Nonrigid medical image registration between images that are linked by an invertible transformation is an inherently symmetric problem. The transformation that registers the image pair should ideally be the inverse of the transformation that registers the pair with the order of images interchanged. This property is referred to as symmetry in registration or inverse consistent registration. However, in practical estimation, the available registration algorithms have tended to produce inverse inconsistent transformations when the template and target images are interchanged. In this paper, we propose two novel cost functions in the large deformation diffeomorphic framework that are inverse consistent. These cost functions have symmetric data-attachment terms; in the first, the matching error is measured at all points along the flow between template and target, and in the second, matching is enforced only at the midpoint of the flow between the template and target. We have implemented these cost functions and present experimental results to validate their inverse consistent property and registration accuracy.
通过可逆变换相联系的图像之间的非刚性医学图像配准本质上是一个对称问题。理想情况下，配准图像对的变换应该是将图像对顺序互换后进行配准的变换的逆变换。这个特性被称为配准中的对称性或逆一致性配准。然而，在实际估计中，当模板图像和目标图像互换时，现有的配准算法往往会产生逆不一致的变换。在本文中，我们在大变形微分同胚框架下提出了两个具有逆一致性的新代价函数。这些代价函数具有对称的数据附着项；在第一个函数中，匹配误差是在模板和目标之间的流的所有点上测量的，而在第二个函数中，匹配仅在模板和目标之间的流的中点处强制进行。我们已经实现了这些代价函数，并给出了实验结果来验证它们的逆一致性特性和配准精度。
