Enhancing electrical impedance tomography reconstruction using learned half-quadratic splitting networks with Anderson acceleration

DOI10.1007/S10915-023-02439-4arXiv2304.14491MaRDI QIDQ6182320FDOQ6182320

Huihui Wang, Qingping Zhou, Guixian Xu

Publication date: 25 January 2024

Published in: Journal of Scientific Computing (Search for Journal in Brave)

Abstract: Electrical Impedance Tomography (EIT) is widely applied in medical diagnosis, industrial inspection, and environmental monitoring. Combining the physical principles of the imaging system with the advantages of data-driven deep learning networks, physics-embedded deep unrolling networks have recently emerged as a promising solution in computational imaging. However, the inherent nonlinear and ill-posed properties of EIT image reconstruction still present challenges to existing methods in terms of accuracy and stability. To tackle this challenge, we propose the learned half-quadratic splitting (HQSNet) algorithm for incorporating physics into learning-based EIT imaging. We then apply Anderson acceleration (AA) to the HQSNet algorithm, denoted as AA-HQSNet, which can be interpreted as AA applied to the Gauss-Newton step and the learned proximal gradient descent step of the HQSNet, respectively. AA is a widely-used technique for accelerating the convergence of fixed-point iterative algorithms and has gained significant interest in numerical optimization and machine learning. However, the technique has received little attention in the inverse problems community thus far. Employing AA enhances the convergence rate compared to the standard HQSNet while simultaneously avoiding artifacts in the reconstructions. Lastly, we conduct rigorous numerical and visual experiments to show that the AA module strengthens the HQSNet, leading to robust, accurate, and considerably superior reconstructions compared to state-of-the-art methods. Our Anderson acceleration scheme to enhance HQSNet is generic and can be applied to improve the performance of various physics-embedded deep learning methods.

Full work available at URL: https://arxiv.org/abs/2304.14491

zbMATH Keywords

electrical impedance tomographynonlinear inverse problemsAnderson accelerationhalf-quadratic splittingalgorithm unrolling

Mathematics Subject Classification ID

Artificial neural networks and deep learning (68T07) Computing methodologies for image processing (68U10) Numerical solutions to equations with nonlinear operators (65J15) Numerical methods for integral transforms (65R10) Inverse problems (including inverse scattering) in optics and electromagnetic theory (78A46)

Cites Work

• Distributed Optimization and Statistical Learning via the Alternating Direction Method of Multipliers
• A first-order primal-dual algorithm for convex problems with applications to imaging
• Linear and Nonlinear Inverse Problems with Practical Applications
• An analysis of electrical impedance tomography with applications to Tikhonov regularization
• On the Convergence of Primal-Dual Hybrid Gradient Algorithm
• A Proof That Anderson Acceleration Improves the Convergence Rate in Linearly Converging Fixed-Point Methods (But Not in Those Converging Quadratically)
• Benchmarking results for the Newton-Anderson method
• Learning nonlinear electrical impedance tomography
• A new non-iterative reconstruction method for the electrical impedance tomography problem
• Isotropic and anisotropic total variation regularization in electrical impedance tomography
• Spatially-adaptive variational reconstructions for linear inverse electrical impedance tomography
• Comments on: ``Anderson acceleration, mixing and extrapolation
• Bayesian Imaging with Data-Driven Priors Encoded by Neural Networks
• Anderson acceleration for contractive and noncontractive operators
• Globally Convergent Type-I Anderson Acceleration for Nonsmooth Fixed-Point Iterations
• Title not available (Why is that?)

Cited In (1)

• Deep unrolling networks with recurrent momentum acceleration for nonlinear inverse problems

This page was built for publication: Enhancing electrical impedance tomography reconstruction using learned half-quadratic splitting networks with Anderson acceleration

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6182320)