Sciweavers

Share
ICMCS
2007
IEEE

Accelerating Mutual-Information-Based Linear Registration on the Cell Broadband Engine Processor

9 years 17 days ago
Accelerating Mutual-Information-Based Linear Registration on the Cell Broadband Engine Processor
Emerging multi-core processors are able to accelerate medical imaging applications by exploiting the parallelism available in their algorithms. We have implemented a mutual-informationbased 3D linear registration algorithm on the Cell Broadband Engine™ processor. By exploiting the highly parallel architecture and its high memory bandwidth, our implementation with two CBE processors can register a pair of 256x256x30 3D images in one second. This implementation is significantly faster than a conventional one on a traditional microprocessor or even faster than a previously reported custom-hardware implementation. In addition to parallelizing the code for multiple cores and organizing the data structure for reducing the amount of the memory traffic, it is also critical to optimize the code for the SIMD pipeline structure. We note that code optimization for the SIMD pipeline alone results in a 4.2x-8.7x acceleration for the computation of small kernels. Further, SIMD optimization alone r...
Moriyoshi Ohara, Hangu Yeo, Frank Savino, Giridhar
Added 03 Jun 2010
Updated 03 Jun 2010
Type Conference
Year 2007
Where ICMCS
Authors Moriyoshi Ohara, Hangu Yeo, Frank Savino, Giridharan Iyengar, Leiguang Gong, Hiroshi Inoue, Hideaki Komatsu, Vadim Sheinin, Shahrokh Daijavad
Comments (0)
books