TY - GEN
T1 - A non-rigid registration method for medical volume data using 3D Phase-Only Correlation
AU - Tajima, Yuichiro
AU - Ito, Koichi
AU - Aoki, Takafumi
PY - 2012/12/1
Y1 - 2012/12/1
N2 - Volume registration is an essential technology for comparing medical volume data acquired in different days and for combining different types of volume data from various imaging devices. For accurate comparison, it is necessary to correct non-rigid deformation between volume data, since the complex deformation between medical volume data is observed even if they are taken from the same regions of the subject. This paper proposes a novel non-rigid registration method using 3D Phase-Only Correlation (POC). The proposed method achieves accurate and fast volume registration by employing POC-based block matching (with small volume blocks), which can find the voxel correspondence with sub-voxel accuracy between two volume data. The proposed method exhibits higher accuracy and shorter computation time compared with the conventional method, and is effective even for multimodality cases such as CT-MRI registration.
AB - Volume registration is an essential technology for comparing medical volume data acquired in different days and for combining different types of volume data from various imaging devices. For accurate comparison, it is necessary to correct non-rigid deformation between volume data, since the complex deformation between medical volume data is observed even if they are taken from the same regions of the subject. This paper proposes a novel non-rigid registration method using 3D Phase-Only Correlation (POC). The proposed method achieves accurate and fast volume registration by employing POC-based block matching (with small volume blocks), which can find the voxel correspondence with sub-voxel accuracy between two volume data. The proposed method exhibits higher accuracy and shorter computation time compared with the conventional method, and is effective even for multimodality cases such as CT-MRI registration.
UR - http://www.scopus.com/inward/record.url?scp=84874578431&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84874578431&partnerID=8YFLogxK
M3 - Conference contribution
AN - SCOPUS:84874578431
SN - 9784990644109
T3 - Proceedings - International Conference on Pattern Recognition
SP - 93
EP - 96
BT - ICPR 2012 - 21st International Conference on Pattern Recognition
T2 - 21st International Conference on Pattern Recognition, ICPR 2012
Y2 - 11 November 2012 through 15 November 2012
ER -