By Topic

Fast Isosurface Extraction for Medical Volume Dataset on Cell BE

Sign In

Cookies must be enabled to login.After enabling cookies , please use refresh or reload or ctrl+f5 on the browser for the login options.

Formats Non-Member Member
$31 $13
Learn how you can qualify for the best price for this item!
Become an IEEE Member or Subscribe to
IEEE Xplore for exclusive pricing!
close button

puzzle piece

IEEE membership options for an individual and IEEE Xplore subscriptions for an organization offer the most affordable access to essential journal articles, conference papers, standards, eBooks, and eLearning courses.

Learn more about:

IEEE membership

IEEE Xplore subscriptions

4 Author(s)
Hai Jin ; Services Comput. Technol. & Syst. Lab., Huazhong Univ. of Sci. & Technol., Wuhan, China ; Bo Li ; Ran Zheng ; Qin Zhang

The size of volumetric data generated by medical imaging and scientific simulations is increased significantly due to the dramatic advances in medical imaging modalities and computing technologies. The volumetric data generally need to be visualized and marching cubes algorithm (MC for short) is one of the standard methods of the isosurface extraction for the medical applications. However, MC algorithm requires a large amount of data computing power. The cell broadband engine (Cell for short) processor, which is a typical COTS (commodity off-the-shelf) heterogeneous designed to handle extremely demanding computations, can be used to hasten isosurface extraction in medial application. In this paper, we present a streaming model-based scheme to efficiently map MC algorithm to Cell. Specifically, a block-based filter running on PPE is imposed as a preprocessing stage to avoid unnecessary data transfer and computation, and the MC kernel runs on SPEs as the subsequent stage. Through tuning the size of the block, the workload of PPE and SPE is orchestrated harmoniously. The experimental results demonstrate that overall isosurface extraction speedup of more than 10 times is achieved compared with conventional heavy iron CPUs.

Published in:

Parallel Processing, 2009. ICPP '09. International Conference on

Date of Conference:

22-25 Sept. 2009