摘要
本文就并行计算技术在医学图像处理领域应用的优势和现状,论述了医学图像处理算法的并行实现方法,并在并行计算集群上实现了CT图像Sobel算子边缘检测及基于互信息的单模PET图像配准的MPI并行算法。针对这两种不同应用和不同的机群规模分析了并行计算的性能,说明了并行计算技术在该领域的适用范围。
This paper introduces the advantages and status quo of using the parallel computing technology in the field of medical image processing, discusses how to achieve parallel algorithms, and implements the MPI parallel algorithms of the CT image edge detection based on the Sobel operator and the PET image registration based on maximal mutual information on parallel computing clusters. After analyzing the performance of parallel computing in the two different applications under different cluster sizes, the scope of the parallel computing application in the medical image field is shown.
出处
《计算机工程与科学》
CSCD
北大核心
2009年第3期32-34,40,共4页
Computer Engineering & Science
关键词
并行计算
MPI
医学图像处理
parallel computing
MPI
medical image processing
