OptiGAN for Crystal Arrays: Physics-Informed Generative Modeling of Optical Photon Transport in PET Detector Arrays

TL;DR

This paper introduces optiGAN, a physics-informed generative model that efficiently simulates optical photon transport in PET detector arrays, significantly reducing computational costs while maintaining high accuracy and generalization capabilities.

Contribution

We develop optiGAN, an enhanced physics-informed GAN that extends single-crystal modeling to array-level simulations with improved generalization and reduced training data requirements.

Findings

Achieves high similarity to Monte Carlo simulations within 3σ agreement.

Successfully generalizes from fundamental domain to full array geometry.

Reproduces characteristic flood map patterns including photopeak clusters.

Abstract

Monte Carlo simulations of optical photon transport are computationally prohibitive for large-scale optical systems including detector arrays and PET systems, restricting practical use to single-crystal studies. This work presents an enhanced conditional generative adversarial network (optiGAN) replacing optical simulations at the crystal array level, extending our single-crystal approach to a 3x3 BGO array. We enhance the Wasserstein-GAN framework with Fourier feature encoding, a learnable latent mapping network, and a physics-informed loss enforcing momentum conservation. Training data is reduced eight-fold by exploiting symmetry. Evaluation employs three studies: a full array evaluation testing generalization from the fundamental domain to the complete geometry, a high-resolution study probing out-of-distribution generalization to untrained positions, and a pencil beam $\gamma$-photon study assessing practical applicability for experimental detector characterization. Performance is benchmarked against GATE10/Geant4 ground truth, using intrinsic fluctuations between independent Monte Carlo runs as baseline. OptiGAN achieves sliced Wasserstein similarity within 3$\sigma$-agreement of the baseline across all conditions, demonstrating successful generalization to the full array. The model transitions from electron-emission training data to realistic $\gamma$-photon interactions, producing flood maps that reproduce characteristic patterns including photopeak clusters and inter-crystal scatter lines. This proof-of-concept demonstrates that physics-informed generative models can accurately simulate optical photon transport in segmented scintillator arrays. The reproduction of experimentally relevant flood map features validates optiGAN for PET detector development and establishes a foundation for models generalizing across diverse array configurations.