Stability of Image-Reconstruction Algorithms

—Robustness and stability of image-reconstruction algorithms have recently come under scrutiny. Their importance to medical imaging cannot be overstated. We review the known results for the topical variational regularization strategies ( ℓ 2 and ℓ 1 regularization) and present novel stability results for ℓ p regularized linear inverse problems for p ∈ (1 , ∞ ) . Our results guarantee Lipschitz continuity for small p and H¨older continuity for larger p . They generalize well to the L p (Ω) function spaces.

[1]  Jan MacDonald,et al.  Solving Inverse Problems With Deep Neural Networks - Robustness Included? , 2020, ArXiv.

[2]  Michael Unser,et al.  Convex Optimization in Sums of Banach Spaces , 2021, 2104.13127.

[3]  L. Rosasco,et al.  Understanding neural networks with reproducing kernel Banach spaces , 2021, Applied and Computational Harmonic Analysis.

[4]  Sebastian Neumayer,et al.  Template based image reconstruction facing different topologies , 2021, 2108.13777.

[5]  Mohammad Zalbagi Darestani,et al.  Measuring Robustness in Deep Learning Based Compressive Sensing , 2021, ICML.

[6]  A. Maleki,et al.  Which bridge estimator is the best for variable selection? , 2020 .

[7]  S. Neumayer,et al.  Stabilizing invertible neural networks using mixture models , 2020, Inverse Problems.

[8]  Michael Unser,et al.  Continuous-Domain Signal Reconstruction Using $L_{p}$-Norm Regularization , 2020, IEEE Transactions on Signal Processing.

[9]  Ben Adcock,et al.  The troublesome kernel: why deep learning for inverse problems is typically unstable , 2020, ArXiv.

[10]  Michael Unser,et al.  A Unifying Representer Theorem for Inverse Problems and Machine Learning , 2019, Foundations of Computational Mathematics.

[11]  Francesco Renna,et al.  On instabilities of deep learning in image reconstruction and the potential costs of AI , 2019, Proceedings of the National Academy of Sciences.

[12]  Michael T. McCann,et al.  Biomedical Image Reconstruction: From the Foundations to Deep Neural Networks , 2019, Found. Trends Signal Process..

[13]  Carola-Bibiane Schönlieb,et al.  Template-Based Image Reconstruction from Sparse Tomographic Data , 2018, Applied Mathematics & Optimization.

[14]  Vidit Saxena,et al.  Spotnet-Learned Iterations For Cell Detection In Image-Based Immunoassays , 2018, 2019 IEEE 16th International Symposium on Biomedical Imaging (ISBI 2019).

[15]  K. Bredies,et al.  Sparsity of solutions for variational inverse problems with finite-dimensional data , 2018, Calculus of Variations and Partial Differential Equations.

[16]  Antonin Chambolle,et al.  On Representer Theorems and Convex Regularization , 2018, SIAM J. Optim..

[17]  Gabriele Steidl,et al.  Regularization of inverse problems via time discrete geodesics in image spaces , 2018, Inverse Problems.

[18]  Michael Unser,et al.  A representer theorem for deep neural networks , 2018, J. Mach. Learn. Res..

[19]  Michael Unser,et al.  Continuous-Domain Solutions of Linear Inverse Problems With Tikhonov Versus Generalized TV Regularization , 2018, IEEE Transactions on Signal Processing.

[20]  Martin Burger,et al.  Modern regularization methods for inverse problems , 2018, Acta Numerica.

[21]  Joakim Jaldén,et al.  Cell Detection by Functional Inverse Diffusion and Nonnegative Group Sparsity—Part I: Modeling and Inverse Problems , 2017, IEEE Transactions on Signal Processing.

[22]  Joakim Jaldén,et al.  Cell Detection by Functional Inverse Diffusion and Non-Negative Group Sparsity—Part II: Proximal Optimization and Performance Evaluation , 2017, IEEE Transactions on Signal Processing.

[23]  P. Weiss,et al.  Exact solutions of infinite dimensional total-variation regularized problems , 2017, Information and Inference: A Journal of the IMA.

[24]  Chong Chen,et al.  Indirect Image Registration with Large Diffeomorphic Deformations , 2017, SIAM J. Imaging Sci..

[25]  Bruce R. Rosen,et al.  Image reconstruction by domain-transform manifold learning , 2017, Nature.

[26]  Michael Unser,et al.  Deep Convolutional Neural Network for Inverse Problems in Imaging , 2016, IEEE Transactions on Image Processing.

[27]  Haolei Weng.  A unified view of high-dimensional bridge regression , 2017 .

[28]  Jens Flemming,et al.  A unified approach to convergence rates for ℓ1-regularization and lacking sparsity , 2015 .

[29]  S. Frick,et al.  Compressed Sensing , 2014, Computer Vision, A Reference Guide.

[30]  Armin Lechleiter,et al.  Tikhonov regularization in Lp applied to inverse medium scattering , 2013 .

[31]  Haizhang Zhang,et al.  Regularized learning in Banach spaces as an optimization problem: representer theorems , 2012, J. Glob. Optim..

[32]  Barbara Kaltenbacher,et al.  Regularization Methods in Banach Spaces , 2012, Radon Series on Computational and Applied Mathematics.

[33]  Ilya Shvartsman.  On stability of minimizers in convex programming , 2012 .

[34]  Shie Mannor,et al.  Sparse Algorithms Are Not Stable: A No-Free-Lunch Theorem , 2008, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[35]  Gilles Blanchard,et al.  On the convergence rate of lp-norm multiple kernel learning , 2012, J. Mach. Learn. Res..

[36]  O. Scherzer,et al.  Necessary and sufficient conditions for linear convergence of ℓ1‐regularization , 2011 .

[37]  T. Poggio,et al.  Sufficient Conditions for Uniform Stability of Regularization Algorithms , 2009 .

[38]  O. Scherzer,et al.  FA ] 2 9 Ju l 2 00 8 Sparse Regularization with l q Penalty Term , 2008 .

[39]  Olivier Gibaru,et al.  C1 and C2-continuous polynomial parametric Lp splines (p>=1) , 2007, Comput. Aided Geom. Des..

[40]  O. Scherzer,et al.  Error estimates for non-quadratic regularization and the relation to enhancement , 2006 .

[41]  M. Nikolova,et al.  Stability of the Minimizers of Least Squares with a Non-Convex Regularization. Part I: Local Behavior , 2006 .

[42]  E. Candès,et al.  Stable signal recovery from incomplete and inaccurate measurements , 2005, math/0503066.

[43]  S. Osher,et al.  Convergence rates of convex variational regularization , 2004 .

[44]  I. Daubechies,et al.  An iterative thresholding algorithm for linear inverse problems with a sparsity constraint , 2003, math/0307152.

[45]  C. Zălinescu.  Convex analysis in general vector spaces , 2002 .

[46]  John E. Lavery.  Univariate cubic Lp splines and shape-preserving, multiscale interpolation by univariate cubic L1 splines , 2000, Comput. Aided Geom. Des..

[47]  Wenjiang J. Fu.  Penalized Regressions: The Bridge versus the Lasso , 1998 .

[48]  A. Goodman,et al.  On uniformly convex functions , 1991 .

[49]  I. Ciorǎnescu.  Geometry of banach spaces, duality mappings, and nonlinear problems , 1990 .

[50]  Mario Bertero,et al.  The Stability of Inverse Problems , 1980 .

[51]  Carl de Boor,et al.  On “best” interpolation☆ , 1976 .