Rigid and Deformable Vasculature-to-Image Registration : a Hierarchical Approach

1
Rigid and Deformable Vasculature-to-Image
Registration a Hierarchical Approach
Julien Jomier and Stephen R. AylwardComputer
Aided-Diagnosis and Display Lab - University of
North Carolina at Chapel Hill

• Introduction
• Combine Rigid Body and Deformable
  Transformations
• Model-to-Image registration
• Hierarchical approach
• No need to extract models of the vessels from
  the target image
• Applications Brain resections, liver tumor
  ablation,

Series of transformations
Match metric

1. Global Rigid Body

2) Piece-wise Rigid following the inherent tree
structure of the vasculature.
3) Deformable registration
Global Rigid
Piece-wise Rigid
Deformable
Deformable Registration
Piece-wise Rigid Transformation
Uses the branch point as starting point.
of centerline points
Rigid body registration applied on each branch
individually in a hierarchical manner.
Distance from centerline (voxel)
ConstraintsRigidity coefficient how much can
we bend the tube? Defined as the maximum angle
allowed between the tangent at rest and the
actual tangent.
Rotation vector for a branch
With ? a linear weighting factor, N the number
of sampling points and the normal vector.
Elasticity coefficient ? how much can we
elongate/shrink the tube to match the image data?
Translation vector
With ? the elasticity coefficient of the parent
branch.
1 Aylward S, Bullitt E, "Initialization, Noise,
Singularities, and Scale in Height-Ridge
Traversal for Tubular Object Centerline
Extraction ," IEEE Transactions on Medical
Imaging, Feb, 2002, Pages 61-75 2 Aylward S,
Jomier J, Weeks S, Bullitt E, "Registration of
Vascular Images," Accepted, International Journal
of Computer Vision, November 2003, pages 15 3
Insight Software Consortium, The insight
toolkit Segmentation and Registration toolkit,
http//www.itk.org
MICCAI - September 2004