Testing deep learning techniques for event reconstruction in pair-production gamma-ray telescopes
Abstract
In the development of next-generation tracking detectors for gamma-ray space observations, the integration of advanced deep learning techniques into event reconstruction algorithms is a promising approach for improving the performance of the mission, reducing systematic uncertainties and maximizing the scientific output. In this study, we investigate the application of deep learning techniques such as Graph Neural Networks (GNNs) for the identification and reconstruction of particle tracks resulting from pair production events within the tracker of the AMEGO-X proposed concept. The goal is improving the angular resolution and the detection efficiency of the telescope, especially in the soft energy range (below 100 MeV) where multiple Coulomb scattering significantly degrades track reconstruction and consequently limits the sensitivity in the so-called MeV gap. We use the Geant4-based MEGAlib framework and a Python-based dedicated read-out and data handler processor to format the input datasets as close as possible to the real data. The simulated datasets are then used to train and evaluate two graph neural network architectures, GraphSAGE and Interaction Networks, and to compare their performance in pair-production event reconstruction in terms of Point Spread Function 68% containment radius and effective area, also with that obtained using standard reconstruction techniques. The results of this study indicate that graph neural network-based reconstruction is a promising approach for pair-production event reconstruction below 100 MeV when compared with standard reconstruction methods, with significant potential for further optimization.
Versions
Related
- The Analysis, not the Aperture: End-to-End Transformer Reconstruction for Imaging Atmospheric Cherenkov Telescopes
- Evolutionary Map of the Universe (EMU): Selection of active galactic nuclei and star-formation emission with machine-learning techniques
- Electric Potential Patterns Forecasting in the Southern Hemisphere with Deep Learning Techniques
- Evolutionary Map of the Universe (EMU): Selection of active galactic nuclei and star-formation emission with machine-learning techniques
- Early Optical Follow-up of Gamma-Ray Bursts: The Critical Role of Robotic Telescopes
BibTeX
@misc{maniscalco2026testing,
title = {{Testing deep learning techniques for event reconstruction in pair-production gamma-ray telescopes}},
author = {Mattia Maniscalco and Valentina Fioretti and Nicolò Parmiggiani and Andrea Bulgarelli and Carolyn A. Kierans and Adrien Laviron and Gabriele Panebianco and Alessio Aboudan and Luca Castaldini and Andreas Zoglauer},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.26350},
eprint = {2609.26350},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.26350},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Testing deep learning techniques for event reconstruction in pair-production gamma-ray telescopes
AU - Mattia Maniscalco
AU - Valentina Fioretti
AU - Nicolò Parmiggiani
AU - Andrea Bulgarelli
AU - Carolyn A. Kierans
AU - Adrien Laviron
AU - Gabriele Panebianco
AU - Alessio Aboudan
AU - Luca Castaldini
AU - Andreas Zoglauer
PY - 2026
DA - 2026/09/22
PB - arXiv
DO - 10.48550/arxiv.2609.26350
UR - https://arxiv.org/abs/2609.26350
AB - In the development of next-generation tracking detectors for gamma-ray space observations, the integration of advanced deep learning techniques into event reconstruction algorithms is a promising approach for improving the performance of the mission, reducing systematic uncertainties and maximizing the scientific output. In this study, we investigate the application of deep learning techniques such as Graph Neural Networks (GNNs) for the identification and reconstruction of particle tracks resulting from pair production events within the tracker of the AMEGO-X proposed concept. The goal is improving the angular resolution and the detection efficiency of the telescope, especially in the soft energy range (below 100 MeV) where multiple Coulomb scattering significantly degrades track reconstruction and consequently limits the sensitivity in the so-called MeV gap. We use the Geant4-based MEGAlib framework and a Python-based dedicated read-out and data handler processor to format the input datasets as close as possible to the real data. The simulated datasets are then used to train and evaluate two graph neural network architectures, GraphSAGE and Interaction Networks, and to compare their performance in pair-production event reconstruction in terms of Point Spread Function 68% containment radius and effective area, also with that obtained using standard reconstruction techniques. The results of this study indicate that graph neural network-based reconstruction is a promising approach for pair-production event reconstruction below 100 MeV when compared with standard reconstruction methods, with significant potential for further optimization.
ER -