An Inexact Newton--Krylov Algorithm for Constrained Diffeomorphic Image Registration

DOI10.1137/140984002MaRDI QIDQ5250023zbMATH OpenOpenAlexWikidataFDO

Authors Andreas Mang, George Biros

Publication date 15 May 2015

Published in SIAM Journal on Imaging Sciences (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1408.6299, http://europepmc.org/articles/pmc5014413

zbMATH Keywords

algorithmcomputational efficiencyconvergencehyperbolic transport equationlarge deformation diffeomorphic image registrationNewton-Krylov methodoptimal controlPDE constrained optimizationPicard methodpreconditioningpseudospectral Galerkin methodStokes regularizationStokes solverTikhonov regularization

Mathematics Subject Classification ID

Existence theories for optimal control problems involving partial differential equations (49J20) Newton-type methods (49M15) Numerical aspects of computer graphics, image analysis, and computational geometry (65D18) Preconditioners for iterative methods (65F08) Numerical optimization and variational techniques (65K10) Image processing (compression, reconstruction, etc.) in information and communication theory (94A08)

Abstract: We propose numerical algorithms for solving large deformation diffeomorphic image registration problems. We formulate the nonrigid image registration problem as a problem of optimal control. This leads to an infinite-dimensional partial differential equation (PDE) constrained optimization problem. The PDE constraint consists, in its simplest form, of a hyperbolic transport equation for the evolution of the image intensity. The control variable is the velocity field. Tikhonov regularization on the control ensures well-posedness. We consider standard smoothness regularization based on $H^1$- or $H^2$-seminorms. We augment this regularization scheme with a constraint on the divergence of the velocity field rendering the deformation incompressible and thus ensuring that the determinant of the deformation gradient is equal to one, up to the numerical error. We use a Fourier pseudospectral discretization in space and a Chebyshev pseudospectral discretization in time. We use a preconditioned, globalized, matrix-free, inexact Newton-Krylov method for numerical optimization. A parameter continuation is designed to estimate an optimal regularization parameter. Regularity is ensured by controlling the geometric properties of the deformation field. Overall, we arrive at a black-box solver. We study spectral properties of the Hessian, grid convergence, numerical accuracy, computational efficiency, and deformation regularity of our scheme. We compare the designed Newton-Krylov methods with a globalized preconditioned gradient descent. We study the influence of a varying number of unknowns in time. The reported results demonstrate excellent numerical accuracy, guaranteed local deformation regularity, and computational efficiency with an optional control on local mass conservation. The Newton-Krylov methods clearly outperform the Picard method if high accuracy of the inversion is required.

Recommendations

• A semi-Lagrangian two-level preconditioned Newton-Krylov solver for constrained diffeomorphic image registration
• A Lagrangian Gauss-Newton-Krylov solver for mass- and intensity-preserving diffeomorphic image registration
• Constrained H^1-regularization schemes for diffeomorphic image registration
• scientific article; zbMATH DE number 7233923
• Numerical optimization for constrained image registration
• A computational framework for image-based constrained registration
• CONFORMAL IMAGE REGISTRATION BASED ON CONSTRAINED OPTIMIZATION
• A novel diffeomorphic model for image registration and its algorithm

Cites work

• A hyperelastic regularization energy for image registration
• A Multigrid Method for a Fourth-Order Diffusion Equation with Application to Image Processing
• A multilevel algorithm for inverse problems with elliptic PDE constraints
• A Multilevel Method for Image Registration
• A Nonlinear Elastic Shape Averaging Approach
• A preconditioning technique for a class of PDE-constrained optimization problems
• A Variational Approach to Nonrigid Morphological Image Registration
• An Inexact SQP Method for Equality Constrained Optimization
• An introduction to continuum mechanics
• An optimal control formulation of an image registration problem
• Analysis of Finite Element Discretizations of an Optimal Control Formulation of the Image Registration Problem
• Brain–Tumor Interaction Biophysical Models for Medical Image Registration
• Choosing the Forcing Terms in an Inexact Newton Method
• Computational Methods for Inverse Problems
• Computing large deformation metric mappings via geodesic flows of diffeomorphisms
• Curvature based image registration
• Determining optical flow
• Diffeomorphic 3D image registration via geodesic shooting using an efficient adjoint calculation
• FAIR. Flexible algorithms for image registration.
• scientific article; zbMATH DE number 1943007 (Why is no real title available?)
• scientific article; zbMATH DE number 2017947 (Why is no real title available?)
• scientific article; zbMATH DE number 2021122 (Why is no real title available?)
• scientific article; zbMATH DE number 3793774 (Why is no real title available?)
• scientific article; zbMATH DE number 5060482 (Why is no real title available?)
• scientific article; zbMATH DE number 961607 (Why is no real title available?)
• Ill-posed medicine—an introduction to image registration
• Image Sequence Interpolation Based on Optical Flow, Segmentation, and Optimal Control
• Image sequence interpolation using optimal control
• Inexact Newton Methods
• Jacobi fields in groups of diffeomorphisms and applications
• Line search multilevel optimization as computational methods for dense optical flow
• Mixed and Hybrid Finite Element Methods
• Nonlinear multigrid methods for total variation image denoising
• Numerical methods for volume preserving image registration
• Numerical optimization for constrained image registration
• Numerical solution of saddle point problems
• On optimization techniques for solving nonlinear inverse problems
• On the application of the Monge-Kantorovich problem to image registration
• Optimal Control Formulation for Determining Optical Flow
• Parallel Lagrange--Newton--Krylov--Schur Methods for PDE-Constrained Optimization. Part I: The Krylov--Schur Solver
• Parallel Lagrange--Newton--Krylov--Schur Methods for PDE-Constrained Optimization. Part II: The Lagrange--Newton Solver and Its Application to Optimal Control of Steady Viscous Flows
• Perspectives in Flow Control and Optimization
• Preconditioned all-at-once methods for large, sparse parameter estimation problems
• Registration of anatomical images using paths of diffeomorphisms parameterized with stationary vector field flows
• Truncated-Newton algorithms for large-scale unconstrained optimization
• Variational problems on flows of diffeomorphisms for image matching

Cited in

(30)

• An augmented Lagrangian method for solving a new variational model based on gradients similarity measures and high order regulariation for multimodality registration
• Fluid image registration using a finite volume scheme of the incompressible Navier Stokes equation
• Coupling brain-tumor biophysical models and diffeomorphic image registration
• Fast iterative solvers for an optimal transport problem
• A lattice Boltzmann method applied to the fluid image registration
• A new multigrid method for unconstrained parabolic optimal control problems
• Diffeomorphic shape matching by operator splitting in 3D cardiology imaging
• Combining the band-limited parameterization and semi-Lagrangian Runge-Kutta integration for efficient PDE-constrained LDDMM
• PDE-constrained optimization in medical image analysis
• Multi-modality image registration models and efficient algorithms
• Fast diffeomorphic image registration via Fourier-approximated Lie algebras
• Algorithm for the reconstruction of dynamic objects in CT-scanning using optical flow
• Analysis of Finite Element Discretizations of an Optimal Control Formulation of the Image Registration Problem
• Constrained H^1-regularization schemes for diffeomorphic image registration
• scientific article; zbMATH DE number 7233923 (Why is no real title available?)
• Improving the Gauss–Newton convergence of a certain position registration scheme
• A semi-Lagrangian two-level preconditioned Newton-Krylov solver for constrained diffeomorphic image registration
• CONFORMAL IMAGE REGISTRATION BASED ON CONSTRAINED OPTIMIZATION
• Multiscale approach for two-dimensional diffeomorphic image registration
• Image-Driven Biophysical Tumor Growth Model Calibration
• A Nash game based variational model for joint image intensity correction and registration to deal with varying illumination
• CLAIRE: a distributed-memory solver for constrained large deformation diffeomorphic image registration
• A Lagrangian Gauss-Newton-Krylov solver for mass- and intensity-preserving diffeomorphic image registration
• An operator-splitting approach for variational optimal control formulations for diffeomorphic shape matching
• AN IMPROVED SIMPLEC SCHEME FOR FLUID REGISTRATION
• A Comparative Study of Different Variants of Newton--Krylov PDE-Constrained Stokes-LDDMM Parameterized in the Space of Band-Limited Vector Fields
• Wavelet-based multiscale flow for realistic image deformation in the large diffeomorphic deformation model framework
• Medical image registration using optimal control of a linear hyperbolic transport equation with a DG discretization
• A hybrid L-BFGS and Gauss-Newton approach for multimodal diffeomorphic image registration
• Higher-order spatial accuracy in diffeomorphic image registration

Describes a project that uses

Uses Software

• iLogDemons
• FAIR.m

This page was built for publication: An Inexact Newton--Krylov Algorithm for Constrained Diffeomorphic Image Registration

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5250023)