Title
Sparse deformation prediction using Markove Decision Processes (MDP) for Non-rigid registration of MR image.
Abstract
•Based on the training image set, a rapid and accurate non-rigid registration is proposed.•The proposed method combines the constructions of Minimum Spanning Tree (MST) and Markove Decision Processes (MDP) at patch scale to provide a suitable initial deformation for the registration from test to template images.•MST provides the registration paths which decomposes the large-deformation registration between the test and the template images into multiple fast registrations with small deformations.•MDP selects the optimal registration path which achieves the most accurate registration.•The proposed method rapidly registers the inter-subject brains and achieves the high mean Dice for the different tissues of the brain.
Year
DOI
Venue
2018
10.1016/j.cmpb.2018.04.024
Computer Methods and Programs in Biomedicine
Keywords
Field
DocType
Deformation prediction,Markove decision processes,Patch-wise registration,MR image
Computer vision,Linkage (mechanical),Computer science,Interpolation,Image scale,Artificial intelligence,Deformation (mechanics),Decision process,Dice,Standard test image,Computation
Journal
Volume
ISSN
Citations 
162
0169-2607
0
PageRank 
References 
Authors
0.34
22
9
Name
Order
Citations
PageRank
Tianyu Fu103.04
qin li201.69
Jianjun Zhu353.10
Danni Ai44514.78
yong huang575.23
Hong Song688.34
Yurong Jiang712611.36
Yongtian Wang845673.00
Jian Yang928348.62