TY - GEN
T1 - Attenuation and Scattering Correction in Passive Gamma Emission Tomography Reconstruction
AU - Fang, Ming
AU - Altmann, Yoann
AU - Della Latta, Daniele
AU - Salvatori, Massimiliano
AU - Di Fulvio, Angela
N1 - Publisher Copyright:
© 2020 IEEE
PY - 2020
Y1 - 2020
N2 - The International Atomic Energy Agency has developed a novel non-destruction assay technique, Passive Gamma Emission Tomography, to inspect spent nuclear fuel assemblies. Accurate physical modeling of the tomography system is required for quantitative tomography reconstruction of the fuel assembly. We have implemented an accelerated Monte Carlo algorithm to compute the imaging system response matrix, applying the attenuation and scattering correction. We performed iterative reconstruction of four simulated fuel assemblies by solving the corresponding imaging inverse problem. The images yielded by the inverse approach were then processed by a convolutional neural network for pin identification. Perfect identification of fuel pins was achieved for medium and high activity level pins. The accuracy of fuel pin activity estimation was significantly improved, compared to the standard filtered back projection (FBP) approach.
AB - The International Atomic Energy Agency has developed a novel non-destruction assay technique, Passive Gamma Emission Tomography, to inspect spent nuclear fuel assemblies. Accurate physical modeling of the tomography system is required for quantitative tomography reconstruction of the fuel assembly. We have implemented an accelerated Monte Carlo algorithm to compute the imaging system response matrix, applying the attenuation and scattering correction. We performed iterative reconstruction of four simulated fuel assemblies by solving the corresponding imaging inverse problem. The images yielded by the inverse approach were then processed by a convolutional neural network for pin identification. Perfect identification of fuel pins was achieved for medium and high activity level pins. The accuracy of fuel pin activity estimation was significantly improved, compared to the standard filtered back projection (FBP) approach.
UR - http://www.scopus.com/inward/record.url?scp=85124688054&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85124688054&partnerID=8YFLogxK
U2 - 10.1109/NSS/MIC42677.2020.9508090
DO - 10.1109/NSS/MIC42677.2020.9508090
M3 - Conference contribution
AN - SCOPUS:85124688054
T3 - 2020 IEEE Nuclear Science Symposium and Medical Imaging Conference, NSS/MIC 2020
BT - 2020 IEEE Nuclear Science Symposium and Medical Imaging Conference, NSS/MIC 2020
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2020 IEEE Nuclear Science Symposium and Medical Imaging Conference, NSS/MIC 2020
Y2 - 31 October 2020 through 7 November 2020
ER -