基于CUDA的数字重建影像生成算法

Generation Algorithm of Digital Reconstruction Radiographs Based on CUDA

鉴于数字重建影像生成过程具有良好的并行性,实现了一种基于CUDA并行计算的数字重建影像生成算法。该算法首先在CPU端使用八叉树结构来剔除体数据中的空体素并将其载入GPU;然后在GPU中根据光线和线程的对应关系,设计光线内核函数来模拟一束X线穿透人体组织的衰减过程;最后在GPU中由多线程并行执行内核函数来完成DRR图像生成过程。实验结果表明,该方法在保证DRR生成质量的前提下能有效利用GPU的并行计算能力,提高DRR图像的生成效率,满足图像引导放疗中对DRR生成过程的实时性要求。

Because the generation procedure of digitally reconstructed radiograph has good parallelism,the digital reconstruction radiograph generation algorithm based on CUDA parallel computing was presented in this paper.Firstly,the octree structure is adopted to organize the volume data in CPU,and then the volume data are loaded into the GPU.The kernel function which can be used to simulate the decay process of X-rays penetrating the human body is designed according to the correspondence between the light and the thread,and finally the kernel function is executed in parallel by the multi-thread to complete the DRR image generation process.The experimental results show that this algorithm uses effectively the parallel computing capabilities of GPU in the premise of ensuring the quality of the DRR,significantly improves the generation speed of DRR,and meets the real-time requirements of DRR in the image-guided radiotherapy.