Robust Model-Based Iteration for Passive Gamma Emission Tomography

TL;DR

This paper introduces an accelerated iterative method combining traditional algorithms with learned operators to improve passive gamma emission tomography reconstructions, reducing computation time while maintaining accuracy.

Contribution

It proposes a novel hybrid solver that integrates deep learning with classical iterative algorithms, ensuring faster convergence without sacrificing reliability.

Findings

The hybrid method achieves similar accuracy to LM in about one third of the iterations.

Different neural architectures exhibit varying robustness to out-of-distribution data.

Experiments on real measurements validate the effectiveness of the proposed approach.

Abstract

Passive Gamma Emission Tomography (PGET) is an IAEA-approved technique for verifying spent nuclear fuel assemblies prior to geological disposal. Reconstructing the emission and attenuation maps from PGET measurements is a nonlinear ill-posed inverse problem, currently solved with a Levenberg-Marquardt (LM) scheme that requires 10-20 iterations to achieve sufficient accuracy. We propose an accelerated iterative solver that combines the LM algorithm with a Deep Gauss-Newton step, in which a learned operator refines the update proposed by the deterministic algorithm at each iteration. A safeguard condition based on the trust-region model ensures that the accelerated iterates perform no worse than LM and retain convergence to a critical point of the regularized objective. Within this framework we compare three architectures for the learned component: an encoder-decoder-style convolutional neural network, Fourier Neural Operators, and Wavelet Neural Operators. Each is trained on a small set of coarsely simulated 9x9 assemblies. Experiments on simulated and real measurements from Finnish nuclear power plants show that the proposed scheme reaches LM-quality reconstructions in roughly one third of the iterations, while revealing architecture-dependent trade-offs in robustness against out-of-distribution inputs.