Generative Prior-Guided Neural Interface Reconstruction for 3D Electrical Impedance Tomography

TL;DR

This paper introduces a hybrid neural framework that combines a pre-trained 3D generative prior with a boundary integral equation solver to improve 3D interface reconstruction in electrical impedance tomography, ensuring physical consistency and data efficiency.

Contribution

It proposes a solver-in-the-loop approach that enforces PDE constraints strictly while leveraging learned shape priors, advancing the state-of-the-art in physics-constrained geometric inverse problems.

Findings

Achieves superior geometric accuracy in 3D EIT reconstructions.

Reduces data requirements compared to traditional methods.

Demonstrates fast, stable convergence with a hybrid physics-prior approach.

Abstract

Reconstructing complex 3D interfaces from indirect measurements remains a grand challenge in scientific computing, particularly for ill-posed inverse problems like Electrical Impedance Tomography (EIT). Traditional shape optimization struggles with topological changes and regularization tuning, while emerging deep learning approaches often compromise physical fidelity or require prohibitive amounts of paired training data. We present a transformative ``solver-in-the-loop'' framework that bridges this divide by coupling a pre-trained 3D generative prior with a rigorous boundary integral equation (BIE) solver. Unlike Physics-Informed Neural Networks (PINNs) that treat physics as soft constraints, our architecture enforces the governing elliptic PDE as a hard constraint at every optimization step, ensuring strict physical consistency. Simultaneously, we navigate a compact latent manifold of plausible geometries learned by a differentiable neural shape representation, effectively regularizing the ill-posed problem through data-driven priors rather than heuristic smoothing. By propagating adjoint shape derivatives directly through the neural decoder, we achieve fast, stable convergence with dramatically reduced degrees of freedom. Extensive experiments on 3D high-contrast EIT demonstrate that this principled hybrid approach yields superior geometric accuracy and data efficiency which is difficult to achieve using traditional methods, establishing a robust new paradigm for physics-constrained geometric discovery.