A Newton method for reconstructing non star-shaped domains in electrical impedance tomography
boundary element methodelectrical impedance tomographyerror estimatesinverse obstacle problemsLaplace equationNewton methodnonlinear inverse problemsnumerical resultsregularizationshape reconstruction
Numerical solutions to equations with nonlinear operators (65J15) Numerical solution to inverse problems in abstract spaces (65J22) Numerical methods for inverse problems for boundary value problems involving PDEs (65N21) Boundary element methods for boundary value problems involving PDEs (65N38) Biological applications of optics and electromagnetic theory (78A70) Biomedical imaging and signal processing (92C55)
The article deals with the problem of the reconstruction of the shape of a perfectly conducting inclusion in three dimensional electrical impedance tomography. This problem is described as the overdetermined boundary value problem \[ \Delta u = 0 \quad \text{in } \Omega, \qquad u = 0 \quad \text{on } \Gamma,\tag{1} \] \[ u = f \quad \text{on } \Sigma,\tag{2} \] \[ \frac{\partial u}{\partial {\mathbf n}} = g \quad \text{on } \Sigma,\tag{3} \] where \(\Omega\) is a given domain with the boundary \(\Sigma\), \(\Gamma\) is an unknown boundary. A parametrization \(\Psi: \Gamma_{\text{ref}} \to \Gamma\), \(\Psi \in H^s(\Gamma_{\text{ref}};{\mathbb R}^3)\), of \(\Gamma\) is considered, and then the operators \(F_v: \Psi \to u\big|_\Sigma\) and \(F_c: \Psi_c: \Psi \to \frac{\partial u}{\partial {\mathbf n}}\big|_\Sigma\) are determined, where \(u\) is the unique solution to the boundary value problem (1), (2) or (1), (3). The inverse problem (1), (2), (3) is equivalent to the operator equation \(F_c(\Psi) = g\) or \(F_v(\Psi) = f\); for studying these equations, Fréchet differentiability of the operators \(F_c\) and \(F_v\) is proved and then some characterizations of the corresponding derivatives and their adjoints are given. At last, for approximate solving the operator equations \(F_c(\Psi) = g\) and \(F_v(\Psi) = f\), a Newton-like method is applied and then, for the numerical solution, the boundary element method. The corresponding error estimates are obtained. In the end of the article numerical results for the reconstruction of a pear-like inclusion and \(H\)-shaped inclusion are presented.
- Recovering boundary shape and conductivity in electrical impedance tomography
- Convergence and regularization for monotonicity-based shape reconstruction in electrical impedance tomography
- A regularized Newton method in electrical impedance tomography using shape Hessian information
- Inclusion boundary reconstruction and sensitivity analysis in electrical impedance tomography
- A new non-iterative reconstruction method for the electrical impedance tomography problem
- A shape decomposition technique in electrical impedance tomography
- On a nonlinear integral equation approach for the surface reconstruction in semi-infinite-layered domains
- A non-iterative sampling approach using noise subspace projection for EIT
- Fréchet derivative with respect to the shape of an internal electrode in electrical impedance tomography
- Higher-order quasi-Monte Carlo for Bayesian shape inversion
- Nonlinear Fourier analysis for discontinuous conductivities: computational results
- Newton regularizations for impedance tomography: convergence by local injectivity
- On Second Order Shape Optimization Methods for Electrical Impedance Tomography
- Shape optimization for 3D electrical impedance tomography
- Comparison of linear and non-linear monotonicity-based shape reconstruction using exact matrix characterizations
- The method of fundamental solutions for complex electrical impedance tomography
- Convergence and regularization for monotonicity-based shape reconstruction in electrical impedance tomography
- Preconditioning of wavelet BEM by the incomplete Cholesky factorization
- Model-aware Newton-type inversion scheme for electrical impedance tomography
- A regularized Newton method in electrical impedance tomography using shape Hessian information
- Monotonicity-based shape reconstruction in electrical impedance tomography
- INTEGRAL REPRESENTATIONS IN ELECTRICAL IMPEDANCE TOMOGRAPHY USING BOUNDARY INTEGRAL OPERATORS
This page was built for publication: A Newton method for reconstructing non star-shaped domains in electrical impedance tomography
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2268303)
