摘要
目的:针对自适应放射治疗中的关键技术--CT和CBCT图像变形配准问题,提出一种基于图形处理器GPU的改进Demons配准算法。方法:通过匹配CT和CBCT图像相应体素局部邻域点集的k阶样本矩,计算CBCT图像每一个体素CT值的线性变换系数,并在每一次Demons迭代过程中,对原CBCT图像逐体素做CT值线性变换,最后利用Demons公式计算变形场。结果:5例临床头颈部肿瘤患者的CT和CBCT图像配准结果表明,改进后算法不受CT和CBCT图像CT值强度不一致的影响,能快速、精确的完成图像的变形配准。结论:基于GPU框架的改进Demons算法可以快速精确完成CT-CBCT图像变形配准,较好的满足了临床对于快速变形图像配准的要求。
Objective：To propose a GPU-based fast CT to Cone-beam CT （CBCT） deformable image registration （DIR） method for adaptive radiotherapy. Methods： Before calculating deformation moving vector field using origin demons, an in- tensity correction step is first performed in CBCT by matching the first and the second moments of the voxel intensities inside a patch around the voxel with those on the CT image. This improved demons algorithm is implanted on the GPU fi＇amework using CUDA language. Results： We use CT and CBCT images ~om five clinical head-and-neck cancer patients to evaluate our algorithm, and compare the DIR results with that of the classic demons algorithm. The results show that our method can register CT and CBCT images with high accuracy in about 80 seconds. Conclusions： We have proposed and evaluated an im- proved demons algorithm for CT-CBCT DIR based on the GPU fi＇amework, and this improved method is a promising tool for clinical adaptive radiotherapy.
出处
《中国医学物理学杂志》
CSCD
2013年第3期4130-4133,共4页
Chinese Journal of Medical Physics
基金
国家自然基金(No.81170078)
广东省科技计划(No.2009B030801360)
广州市科技计划(No.2011J4300131)
