Deep Learning reconstruction with uncertainty estimation for γ photon interaction in fast scintillator detectors - CEA - Commissariat à l’énergie atomique et aux énergies alternatives
Article Dans Une Revue Engineering Applications of Artificial Intelligence Année : 2024

Deep Learning reconstruction with uncertainty estimation for γ photon interaction in fast scintillator detectors

Résumé

This article presents a physics-informed deep learning method for the quantitative estimation of the spatial coordinates of gamma interactions within a monolithic scintillator, with a focus on Positron Emission Tomography (PET) imaging. The authors propose a density neural network approach to estimate the 2-dimensional gamma photon interaction coordinates for a fast lead tungstate (PbWO$_4$) monolithic scintillator detector. This approach considers uncertainties on the estimated coordinates and accounts for physical constraints on the detector. The results demonstrate the effectiveness of the proposed approach for accurate position estimation and the benefits of the uncertainties estimation. The authors discuss its potential impact on improving PET imaging quality and highlight the generalization of the proposed methodology for other applications.
Fichier principal
Vignette du fichier
1-s2.0-S0952197624000344-main.pdf (4.52 Mo) Télécharger le fichier
AAIMME_Paper.pdf (4.67 Mo) Télécharger le fichier
Origine Publication financée par une institution
Licence
Origine Fichiers produits par l'(les) auteur(s)

Dates et versions

cea-04458433 , version 1 (14-02-2024)

Licence

Identifiants

Citer

Geoffrey Daniel, Mohamed-Bahi Yahiaoui, Claude Comtat, Sebastien Jan, Olga Kochebina, et al.. Deep Learning reconstruction with uncertainty estimation for γ photon interaction in fast scintillator detectors. Engineering Applications of Artificial Intelligence, 2024, 131, pp.107876. ⟨10.1016/j.engappai.2024.107876⟩. ⟨cea-04458433⟩
25 Consultations
10 Téléchargements

Altmetric

Partager

More