A primal-dual hybrid gradient method for non-linear operators with applications to MRI

Abstract

We study the solution of minimax problems ${\min }_x{\max }_{y}G(x)+⟨K(x),y⟩-F^{\ast }(y)$ in finite-dimensional Hilbert spaces. The functionals $G$ and $F^{\ast }$ we assume to be convex, but the operator $K$ we allow to be non-linear. We formulate a natural extension of the modified primal-dual hybrid gradient method (PDHGM), originally for linear $K$, due to Chambolle and Pock. We prove the local convergence of the method, provided various technical conditions are satisfied. These include in particular the Aubin property of the inverse a monotone operator at the solution. Of particular interest to us is the case arising from reformulation of regularisation problems ${\min }_x\Vert f-T(x){\Vert }^2/2+\alpha R(x)$ with the operator $T$ non-linear. For such problems, we show that our general local convergence result holds when the noise level of the data $f$ is low, and the regularisation parameter $\alpha$ is correspondingly small. We verify the numerical performance of the method by applying it to problems from magnetic resonance imaging (MRI) in chemical engineering and medicine. The specific applications are in diffusion tensor imaging (DTI) and MR velocity imaging. These numerical studies show very promising performance.

Cited by in corpus (30)

• Learned Primal-dual Reconstruction
• Solving ill-posed inverse problems using iterative deep neural networks
• Variational Depth from Focus Reconstruction
• On the convergence of a linesearch based proximal-gradient method for nonconvex optimization
• Primal-dual extragradient methods for nonlinear nonsmooth PDE-constrained optimization
• The structure of optimal parameters for image restoration problems
• Relaxed Gauss-Newton methods with applications to electrical impedance tomography
• Solving Inverse Problems by Joint Posterior Maximization with Autoencoding Prior
• Preconditioned ADMM with nonlinear operator constraint
• Acceleration and global convergence of a first-order primal--dual method for nonconvex problems
• The jump set under geometric regularisation. Part 1: Basic technique and first-order denoising
• Primal-dual proximal splitting and generalized conjugation in non-smooth non-convex optimization
• First-order primal-dual methods for nonsmooth nonconvex optimisation
• Acceleration of the PDHGM on strongly convex subspaces
• Mumford-Shah regularization in electrical impedance tomography with complete electrode model
• Stability of saddle points via explicit coderivatives of pointwise subdifferentials
• Linearly convergent bilevel optimization with single-step inner methods
• Temporal Huber regularization for DCE-MRI
• An Extended Primal-Dual Algorithm Framework for Nonconvex Problems with Application to Nonlinear Imaging
• Optimal control of the principal coefficient in a scalar wave equation
• Quantitative MR Image Reconstruction using Parameter-Specific Dictionary Learning with Adaptive Dictionary-Size and Sparsity-Level Choice
• An Inexact Semi-smooth Newton Method on Riemannian Manifolds with Application to Duality-based Total Variation Denoising
• Fenchel Duality and a Separation Theorem on Hadamard Manifolds
• Fenchel Duality Theory and A Primal-Dual Algorithm on Riemannian Manifolds
• A nonsmooth primal-dual method with interwoven PDE constraint solver
• First- and Second-Order Analysis for Optimization Problems with Manifold-Valued Constraints
• Diffusion tensor imaging with deterministic error bounds
• Online optimisation for dynamic electrical impedance tomography
• Complex extension of optical flow and its practical evaluation for undersampled dynamic MRI
• A primal-dual splitting algorithm for monotone inclusions with applications