基于CUDA加速的三维医学图像配准

CUDA-based Acceleration of Three Dimensional 3D Medical Images Registration

三维医学图像配准技术是医学图像处理,特别是外科手术导航的关键技术,但现有的三维医学图像配准算法大多存在计算量大、耗时过长的问题,不能满足临床应用中实时处理的要求.针对这一问题,提出一种基于统一计算设备架构(Compute Unified Device Architecture,CUDA)的高性能计算方法,充分利用CUDA架构下GPU(Graphic Processing Unit)并行计算的优势,并结合图像多尺度、最大互信息等方法,实现了三维医学图像的快速配准.实验结果表明,该方法在保证配准精度的前提下,大幅度地提高了三维医学图像配准算法的运算速度,可以满足临床上对配准算法的实时性要求.

Real time 3D medical image registration method is key technology of medical image processing, especially in surgical oper-ation navigation. However, current 3D medical image registration methods are time-consuming, which can＇t meet the real time re-quirement of clinical application. To solve this problem, this paper presented a high performance computational method based on CU- DA （ Compute Unified Device Architecture}, which took full advantage of GPU parallel computing under CUDA architecture com- bined with image multiple scale and maximum mutual information to make fast registration of three dimensional medical image. Ex-periments showed that this algorithm can greatly accelerate the computational speed of rgistration of three dimensional medical im- age, and meet the real time requirement of clinical application.