【24h】

DEFORMABLE REGISTRATION METHOD USING TPS AND B-SPLINE

机译:TPS和B样条的可变形配准方法

获取原文
获取原文并翻译 | 示例
获取外文期刊封面目录资料

摘要

We develop a hybrid method joint using combining Thin-rnPlate Spline(TPS) and B-spline to improve both thernefficiency and accuracy of the conventional B-spline inrnregistration of CT image. The proposed algorithm consistsrnof two steps. First, we use TPS as a global registrationrnmethod to register two CT images rapidly, deformingrnlarge unfit regions in the moving image to matchrncounterpart in the reference image. Then B-spline is usedrnfor local registration, the previous deformed movingrnimage using TPS is further deformed to match thernreference image more accurately. In our experiment, arncouple of clinical images are simulated using the two-steprnalgorithm above, result in a tremendous improvement inrnboth run-time and registration quality, compared with thernconventional methods solely using either TPS or B-spline.rnIt is demonstrated that the proposed method can combinernthe efficiency of TPS and the accuracy of B-spline,rnperforming good adaptivity and robust in registration ofrnclinical CT image.
机译:我们结合使用Thin-rnPlate Spline(TPS)和B样条开发了一种混合方法关节,以提高传统B样条未对准CT图像的效率和准确性。所提出的算法包括两个步骤。首先,我们使用TPS作为全局配准方法来快速配准两个CT图像,使运动图像中的较大不适合区域变形以匹配参考图像中的对应部分。然后将B样条曲线用于局部配准,使用TPS将先前变形的运动图像进一步变形以更精确地匹配参考图像。在我们的实验中,使用上述两步算法模拟临床图像的神经耦合,与仅使用TPS或B样条的常规方法相比,在运行时间和配准质量上都有了极大的改善。rn证明了该方法可以结合TPS的效率和B样条的准确性,在临床CT图像配准方面表现出良好的适应性和鲁棒性。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号