TY - CHAP
T1 - Non-uniform gradient prescription for precise angular measurements using DTI.
AU - Yanasak, Nathan
AU - Allison, Jerry D.
AU - Zhao, Qun
AU - Hu, Tom C C
AU - Dhandapani, Krishnan
PY - 2008/12/1
Y1 - 2008/12/1
N2 - Diffusion Tensor Imaging (DTI) calculates a tensor for each voxel, representing the mean diffusive characteristics in volume-averaged tissue. Gradients that phase-encode spins according to the amount of their diffusion are usually applied uniformly over a sphere during a DTI procedure for minimal bias of tensor information. If prior knowledge of diffusion direction exists, the angular precision for determining the principle eigenvector of cylindrically-symmetric ("prolate") tensors can be improved by specifying gradients non-uniformly. Improvements in precision of 30-40% can be achieved using a restricted band of zenith angle values for gradient directions. Sensitivity to the a priori angular range of the principle eigenvector can be adjusted with the width of the band. Simulations and phantom data are in agreement; a preliminary validation is presented.
AB - Diffusion Tensor Imaging (DTI) calculates a tensor for each voxel, representing the mean diffusive characteristics in volume-averaged tissue. Gradients that phase-encode spins according to the amount of their diffusion are usually applied uniformly over a sphere during a DTI procedure for minimal bias of tensor information. If prior knowledge of diffusion direction exists, the angular precision for determining the principle eigenvector of cylindrically-symmetric ("prolate") tensors can be improved by specifying gradients non-uniformly. Improvements in precision of 30-40% can be achieved using a restricted band of zenith angle values for gradient directions. Sensitivity to the a priori angular range of the principle eigenvector can be adjusted with the width of the band. Simulations and phantom data are in agreement; a preliminary validation is presented.
UR - http://www.scopus.com/inward/record.url?scp=58849086830&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=58849086830&partnerID=8YFLogxK
U2 - 10.1007/978-3-540-85988-8_103
DO - 10.1007/978-3-540-85988-8_103
M3 - Chapter
C2 - 18979827
SN - 354085987X
SN - 9783540859871
VL - 11
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 866
EP - 873
BT - Medical image computing and computer-assisted intervention : MICCAI ... International Conference on Medical Image Computing and Computer-Assisted Intervention
T2 - 11th International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2008
Y2 - 6 September 2008 through 10 September 2008
ER -