Deep Unfolding Network for Nonlinear Multi-Frequency Electrical Impedance Tomography

TL;DR

This paper introduces a deep unfolding network that combines classical iterative methods with graph neural networks to improve nonlinear multi-frequency electrical impedance tomography, enhancing tissue conductivity imaging accuracy.

Contribution

It proposes a novel variational network integrating GNNs within the PRGN framework for better nonlinear model fitting and inter-frequency correlation capture in mfEIT.

Findings

Effective reconstruction of tissue conductivities across frequencies.

Preserves mesh structure for accurate tissue fraction estimation.

Combines interpretability of classical methods with deep learning capabilities.

Abstract

Multi-frequency Electrical Impedance Tomography (mfEIT) represents a promising biomedical imaging modality that enables the estimation of tissue conductivities across a range of frequencies. Addressing this challenge, we present a novel variational network, a model-based learning paradigm that strategically merges the advantages and interpretability of classical iterative reconstruction with the power of deep learning. This approach integrates graph neural networks (GNNs) within the iterative Proximal Regularized Gauss Newton (PRGN) framework. By unrolling the PRGN algorithm, where each iteration corresponds to a network layer, we leverage the physical insights of nonlinear model fitting alongside the GNN's capacity to capture inter-frequency correlations. Notably, the GNN architecture preserves the irregular triangular mesh structure used in the solution of the nonlinear forward model, enabling accurate reconstruction of overlapping tissue fraction concentrations.