A CUDA-based reverse gridding algorithm for MR reconstruction
A CUDA-based reverse gridding algorithm for MR reconstruction
复制标题
基于CUDA的MR重建逆网格算法
DOI：
10.1016/j.mri.2012.06.038
复制
发表时间：
2013-02-01
影响因子：
2.5
通讯作者：
Zhao, Dazhe
中科院分区：
文献类型：
--
作者：
Yang, Jingzhu;Feng, Chaolu;Zhao, Dazhe
使用非笛卡尔方法采集的磁共振原始数据可通过传统的网格化算法（GA）在笛卡尔网格上进行转换，并通过傅里叶变换进行重建。然而，其运行时间复杂度为O(KxN²)，其中原始数据的分辨率为NxN，卷积窗口（CW）的大小为K。并且它涉及大量的矩阵计算，包括求模、加法、乘法和卷积。因此，提出了一种基于计算统一设备架构（CUDA）的算法，以提高螺旋桨（一种全球公认的非笛卡尔采样方法）的重建效率。实验表明在多个CUDA线程之间存在写 - 写冲突。这在将多个k - 空间数据同步卷积到同一网格上时会导致不一致的结果。为了克服这个问题，开发了一种反向网格化算法（RGA）。与传统GA中为每个轨迹生成一个网格窗口的方法不同，RGA为每个网格计算一个轨迹窗口。这就是“反向”的含义。对于CW中的每个k - 空间点，其贡献累加到该网格。尽管该算法可轻松扩展以重建其他非笛卡尔采样的原始数据，但我们仅基于螺旋桨实现它。实验表明，这种基于CUDA的RGA成功解决了写 - 写冲突，并且其重建速度比传统GA高7.5倍。（C）2013爱思唯尔公司。保留所有权利。
MR raw data collected using non-Cartesian method can be transformed on Cartesian grids by traditional gridding algorithm (GA) and reconstructed by Fourier transform. However, its runtime complexity is O(KxN(2)), where resolution of raw data is NxN and size of convolution window (CW) is K. And it involves a large number of matrix calculation including modulus, addition, multiplication and convolution. Therefore, a Compute Unified Device Architecture (CUDA)-based algorithm is proposed to improve the reconstruction efficiency of PROPELLER (a globally recognized non-Cartesian sampling method). Experiment shows a write-write conflict among multiple CUDA threads. This induces an inconsistent result when synchronously convoluting multiple k-space data onto the same grid. To overcome this problem, a reverse gridding algorithm (RGA) was developed. Different from the method of generating a grid window for each trajectory as in traditional GA, RGA calculates a trajectory window for each grid. This is what "reverse" means. For each k-space point in the CW, contribution is cumulated to this grid. Although this algorithm can be easily extended to reconstruct other non-Cartesian sampled raw data, we only implement it based on PROPELLER. Experiment illustrates that this CLTDA-based RGA has successfully solved the write-write conflict and its reconstruction speed is 7.5 times higher than that of traditional GA. (C) 2013 Elsevier Inc. All rights reserved.