High-Performance Deformable Image Registration Algorithms for Manycore Processors by James Shackleford & Nagarajan Kandasamy & Gregory Sharp

Author:James Shackleford & Nagarajan Kandasamy & Gregory Sharp
Language: eng
Format: epub
ISBN: 9780124078802
Publisher: Elsevier Inc.
Published: 2013-07-01T16:00:00+00:00

(3.5)

where N is the total number of voxels mapping from S to M. However, this cost function cannot be used to assess the quality of a deformation field that is attempting to register images acquired using different imaging modalities since these images may have differing voxel-intensity maps for identical anatomy. For such multimodality registrations, the more sophisticated cost function of MI may be used which qualifies the amount of information content the two images share in common; the images will be optimally aligned when the shared information content is maximum. To understand MI as a cost function, consider the intensity a of a voxel located at coordinates within the static image, , and the intensity b of a voxel at coordinates within the moving image, . The goal is to apply a coordinate transform to the moving image such that it registers best with the static image. The statistical MI is obtained as

Download

Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.