Construct Deep Neural Networks based on Direct Sampling Methods for Solving Electrical Impedance Tomography

DOI10.1137/20M1367350zbMath1479.65020arXiv2009.08024OpenAlexW3168992342MaRDI QIDQ4997381

Ruchi Guo, Jiahua Jiang

Publication date: 29 June 2021

Published in: SIAM Journal on Scientific Computing (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/2009.08024

zbMATH Keywords

inverse problemsreconstruction algorithmelectrical impedance tomographydeep learningdirect sampling methodslimited boundary data

Mathematics Subject Classification ID

Computing methodologies for image processing (68U10) Neural biology (92C20) Biomedical imaging and signal processing (92C55) Ill-posed problems for PDEs (35R25) Inverse problems for PDEs (35R30) Numerical methods for inverse problems for boundary value problems involving PDEs (65N21) Inverse problems (including inverse scattering) in optics and electromagnetic theory (78A46) Numerical methods for ill-posed problems for boundary value problems involving PDEs (65N20)

Related Items (6)

Imaging conductivity from current density magnitude using neural networks* ⋮ Electrical impedance tomography with deep Calderón method ⋮ Divide-and-conquer DNN approach for the inverse point source problem using a few single frequency measurements ⋮ Learn an index operator by CNN for solving diffusive optical tomography: a deep direct sampling method ⋮ A direct sampling-based deep learning approach for inverse medium scattering problems ⋮ A Data-Assisted Two-Stage Method for the Inverse Random Source Problem

Uses Software

• DeepLab
• Adam
• CNN-RNN

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• Higher-order topological sensitivity for 2-D potential problems. Application to fast identification of inclusions
• Polarization and moment tensors. With applications to inverse problems and effective medium theory
• Level set and total variation regularization for elliptic inverse problems with discontinuous coefficients.
• Provable approximation properties for deep neural networks
• Electrical impedance tomography using level set representation and total variational regularization
• Reconstruction of small inhomogeneities from boundary measurements
• Approximation rates for neural networks with general activation functions
• A fixed mesh method with immersed finite elements for solving interface inverse problems
• A direct approach for function approximation on data defined manifolds
• Solving electrical impedance tomography with deep learning
• Calderón's inverse conductivity problem in the plane
• Explicit Characterization of Inclusions in Electrical Impedance Tomography
• Level-set function approach to an inverse interface problem
• A reconstruction algorithm for electrical impedance tomography based on sparsity regularization
• A direct sampling method for electrical impedance tomography
• 10.1162/153244304322972667
• Electric Current Computed Tomography and Eigenvalues
• A direct sampling method to an inverse medium scattering problem
• Identification of conductivity imperfections of small diameter by boundary measurements. Continuous dependence and computational reconstruction
• Universal approximation bounds for superpositions of a sigmoidal function
• Enhanced Electrical Impedance Tomographyviathe Mumford–Shah Functional
• An inverse source problem in potential analysis
• Numerical implementation of two noniterative methods for locating inclusions by impedance tomography
• A Time-Dependent Direct Sampling Method for Recovering Moving Potentials in a Heat Equation
• A convergent adaptive finite element method for electrical impedance tomography
• A group of immersed finite-element spaces for elliptic interface problems
• Direct Sampling Method for Diffusive Optical Tomography
• A direct sampling method for inverse electromagnetic medium scattering
• On the inverse conductivity problem with one measurement

This page was built for publication: Construct Deep Neural Networks based on Direct Sampling Methods for Solving Electrical Impedance Tomography