Regularity-guaranteed transformation estimation in medical image registration

Bibo Shi, Jundong Liu, David R. Haynor, Sébastien Ourselin
2012 Medical Imaging 2012: Image Processing  
SHI, BIBO, M.S., August 2011, Biomedical Engineering Regularity-Guaranteed Transformation Estimation in Medical Image Registration (64 pp.) Director of Thesis: Jundong Liu In addition to seeking geometric correspondence between the inputs, a legitimate medical image registration algorithm should also keep the estimated transformation meaningful or regular. In this thesis, we present a mathematically sound formulation that explicitly controls the transformation to keep each grid in a meaningful
more » ... hape over the entire geometric matching procedure. The deformation regularity conditions are enforced by maintaining all the moving neighbors as non-twist grids. In contrast to similar work, we differentiate and formulate the convex and concave folding cases under an efficient and straightforward point-to-line/surface orientation framework, and use equality constraints to guarantee grid regularity and prevent folding. The equality constrained optimization problem is efficiently solved using the augmented Lagrangian Mulplier method. Experiments on human brain MR images are presented to show the improvements made by our model over the popular Demon's and DCT-based registration algorithms. Extension to develop a clinical registration package including the regularity guaranteed conditions is also explored.
doi:10.1117/12.911083 dblp:conf/miip/ShiL12 fatcat:pgftexe7c5bdzbnubdluivyc5e