Rapid rabbit: Highly optimized GPU accelerated cone-beam CT reconstruction
Rapid rabbit: Highly optimized GPU accelerated cone-beam CT reconstruction
复制标题
Rapidrabbit：高度优化的GPU加速锥束CT重建
DOI：
10.1109/nssmic.2013.6829126
复制
发表时间：
2013
期刊：
影响因子：
--
通讯作者：
K. Mueller
中科院分区：
文献类型：
--
作者：
Eric Papenhausen;K. Mueller
图形处理单元（GPU）已经在医学成像领域中被广泛采用。GPU的并行SIMD性质完美地映射到许多重建算法。因此，并行化常见的重建算法（例如FDK反投影）相对简单。这意味着显著的性能改进必须来自仔细的内存优化，利用ASIC和其他一些技巧来提高指令吞吐量。我们提出了优化，建立了以前的工作，以优化GPU加速FDK反投影实现使用RabbitCT数据集。
Graphical processing units (GPUs) have become widely adopted in the medical imaging community. The parallel SIMD nature of GPUs maps perfectly to many reconstruction algorithms. Because of this, it is relatively straightforward to parallelize common reconstruction algorithms (e.g. FDK backprojection). This means that significant performance improvements must come from careful memory optimizations, exploiting ASICs and a few other tricks to boost instruction throughput. We present optimizations that build off of previous work to optimize a GPU accelerated FDK backprojection implementation using the RabbitCT dataset.