3D Convolutional Neural Network to Enhance Small-Animal Positron Emission Tomography Images in the Sinogram Domain

Leandro José Rodríguez Hernández, Humberto de Jesús Ochoa Domínguez, Osslan Osiris Vergara Villegas, Vianey Guadalupe Cruz Sánchez, Juan Humberto Sossa Azuela, Javier Polanco González

Resultado de la investigación: Capítulo del libro/informe/acta de congresoContribución a la conferenciarevisión exhaustiva

Resumen

In this work, we propose a three dimensional (3D) convolutional neural network (CNN) to enhance sinograms acquired from a small-animal positron emission tomography (PET) scanner. The network consists of three convolutional layers created with 3D filters of sizes 9, 3, and 5, respectively. We extracted 15250 3D patches from low- and high-count sinograms to build the low- and high-resolution pairs for training. After training and prediction, the enhanced sinogram is reconstructed using the ordered subset expectation maximization (OSEM) algorithm. The results revealed that the proposed network improved the spillover ratio and the uniformity of the standard NU4-2008 phantom up to 8% and 75%, respectively.

Idioma originalInglés
Título de la publicación alojadaPattern Recognition - 14th Mexican Conference, MCPR 2022, Proceedings
EditoresOsslan Osiris Vergara-Villegas, Vianey Guadalupe Cruz-Sánchez, Juan Humberto Sossa-Azuela, Jesús Ariel Carrasco-Ochoa, José Francisco Martínez-Trinidad, José Arturo Olvera-López
EditorialSpringer Science and Business Media Deutschland GmbH
Páginas94-104
Número de páginas11
ISBN (versión impresa)9783031077494
DOI
EstadoPublicada - 2022
Evento14th Mexican Conference on Pattern Recognition, MCPR 2022 - Ciudad Juárez, México
Duración: 22 jun 202225 jun 2022

Serie de la publicación

NombreLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volumen13264 LNCS
ISSN (versión impresa)0302-9743
ISSN (versión digital)1611-3349

Conferencia

Conferencia14th Mexican Conference on Pattern Recognition, MCPR 2022
País/TerritorioMéxico
CiudadCiudad Juárez
Período22/06/2225/06/22

Huella

Profundice en los temas de investigación de '3D Convolutional Neural Network to Enhance Small-Animal Positron Emission Tomography Images in the Sinogram Domain'. En conjunto forman una huella única.

Citar esto