Motion estimation and correction for myocardial perfusion imaging

Motion estimation and correction is an important yet challenge problem in medical image analysis. This paper presents a new approach to estimate and correct motion artifacts in cardiac perfusion image sequences. The method uses a modified optical flow and block-matching technique to estimate the vel...

Full description

Saved in:
Bibliographic Details
Published inIFAC Proceedings Volumes Vol. 36; no. 15; pp. 157 - 161
Main Authors Hsu, Li-yueh, Lin, Kang-Ping, Hu, Yu-Chang, Yamada, Naoaki, Watabe, Hiroshi, Iida, Hidehiro
Format Journal Article
LanguageEnglish
Published 01.08.2003
Subjects
motion estimation
optical flows
registration
optical flows
motion estimation
registration
Online AccessGet full text

Cover

More Information
Summary:Motion estimation and correction is an important yet challenge problem in medical image analysis. This paper presents a new approach to estimate and correct motion artifacts in cardiac perfusion image sequences. The method uses a modified optical flow and block-matching technique to estimate the velocity vector between consecutive image frames. The displacement field of each image is corrected and deformed according a total velocity. This technique is applied to sequences of magnetic resonance myocardial perfusion images. Results from quantitative error measurement and time-intensity plots show improvements over original images after the use of this technique.
ISSN:1474-6670
DOI:10.1016/S1474-6670(17)33492-4