Objective: To investigate CT image non-rigid registration using the B spline in Radiotherapy. Methods: Using B-spline free-form deformations model and SSD as a similarity measure function, based on multi-resolution B-spline non-rigid registration method to registration CT image with the same patient before and after radiotherapy. Results: The non-rigid registration result is more satisfactory by Gaussian filtering image and using B-spline free-form deformations model. Conclusions: CT image non-rigid registration using the B spline can be modeling deformation of the tumor region, and provide support for clinical analysis of changes in the tumor.%目的:应用B样条弹性模型研究肿瘤患者放疗前CT图像与放疗后CT图像之间的弹性配准.方法:利用B样条自由弹性模型(free-form deformations,FFD),以灰度差平方和(Sum of Squared Differences,SSD)为相似性测度函数,通过基于多分辨率的B样条弹性配准方法对同一患者放疗前与放疗后的CT图像进行配准.结果:图像通过高斯滤波处理,应用B样条弹性模型,得到较为理想的弹性配准实验结果.结论:基于B样条的CT与CT图像弹性配准可以较好地建模放疗前与放疗后肿瘤区域的形变,为临床分析肿瘤的变化提供支持.
展开▼