Towards a Whole Body [<sup>18</sup><i>F</i>] FDG Positron Emission Tomography Attenuation Correction Map Synthesizing using Deep Neural Networks

The correction of attenuation effects in Positron Emission Tomography (PET) imaging is fundamental to obtain a correct radiotracer distribution. However direct measurement of this attenuation map is not error-free and normally results in additional ionization radiation dose to the patient. Here, we...

Descripción completa

Guardado en:
Detalles Bibliográficos
Autores principales: Rodríguez Colmeiro, Ramiro, Verrastro, Claudio, Minsky, Daniel, Grosges, Thomas
Formato: Articulo
Lenguaje:Inglés
Publicado: 2021
Materias:
Ciencias Informáticas
Attenuation Correction
Deep Learning
Generative Models
Positron Emission Tomography
Corrección de atenuación
Aprendizaje profundo
Modelos generativos
Tomografía por Emisión de Positrones
Acceso en línea:http://sedici.unlp.edu.ar/handle/10915/118045
https://journal.info.unlp.edu.ar/JCST/article/view/1519
Aporte de:
SEDICI (UNLP) de Universidad Nacional de La Plata
id I19-R120-10915-118045
record_format dspace
institution Universidad Nacional de La Plata
institution_str I-19
repository_str R-120
collection SEDICI (UNLP)
language Inglés
topic Ciencias Informáticas
Attenuation Correction
Deep Learning
Generative Models
Positron Emission Tomography
Corrección de atenuación
Aprendizaje profundo
Modelos generativos
Tomografía por Emisión de Positrones
spellingShingle Ciencias Informáticas
Attenuation Correction
Deep Learning
Generative Models
Positron Emission Tomography
Corrección de atenuación
Aprendizaje profundo
Modelos generativos
Tomografía por Emisión de Positrones
Rodríguez Colmeiro, Ramiro
Verrastro, Claudio
Minsky, Daniel
Grosges, Thomas
Towards a Whole Body [<sup>18</sup><i>F</i>] FDG Positron Emission Tomography Attenuation Correction Map Synthesizing using Deep Neural Networks
topic_facet Ciencias Informáticas
Attenuation Correction
Deep Learning
Generative Models
Positron Emission Tomography
Corrección de atenuación
Aprendizaje profundo
Modelos generativos
Tomografía por Emisión de Positrones
description The correction of attenuation effects in Positron Emission Tomography (PET) imaging is fundamental to obtain a correct radiotracer distribution. However direct measurement of this attenuation map is not error-free and normally results in additional ionization radiation dose to the patient. Here, we explore the task of whole body attenuation map generation using 3D deep neural networks. We analyze the advantages that an adversarial training can provide to such models. The networks are trained to learn the mapping from non attenuation corrected [<sup> 18</sup><i>F</i>]-fluorodeoxyglucose PET images to a synthetic Computerized Tomography (sCT) and also to label the input voxel tissue. Then the sCT image is further refined using an adversarial training scheme to recover higher frequency details and lost structures using context information. This work is trained and tested on public available datasets, containing several PET images from different scanners with different radiotracer administration and reconstruction modalities. The network is trained with 108 samples and validated on 10 samples. The sCT generation was tested on 133 samples from 8 distinct datasets. The resulting mean absolute error of the tested networks is 96 ± 20 HU and 103 ± 18 HU with a peak signal to noise ratio of 19.3 ± 1.7 dB and 18.6 ± 1.5 dB, for the base model and adversarial model respectively. The attenuation correction is tested by means of attenuation sinograms, obtaining a line of response attenuation mean error lower than 1% with a standard deviation lower than 8%. The proposed deep learning topologies are capable of generating whole body attenuation maps from uncorrected PET image data. Moreover, the accuracy of both methods holds in the presence of data from multiple sources and modalities and are trained on publicly available datasets. Finally, while the adversarial layer enhances visual appearance of the produced samples, the 3D U-Net achieves higher metric performance.
format Articulo
Articulo
author Rodríguez Colmeiro, Ramiro
Verrastro, Claudio
Minsky, Daniel
Grosges, Thomas
author_facet Rodríguez Colmeiro, Ramiro
Verrastro, Claudio
Minsky, Daniel
Grosges, Thomas
author_sort Rodríguez Colmeiro, Ramiro
title Towards a Whole Body [<sup>18</sup><i>F</i>] FDG Positron Emission Tomography Attenuation Correction Map Synthesizing using Deep Neural Networks
title_short Towards a Whole Body [<sup>18</sup><i>F</i>] FDG Positron Emission Tomography Attenuation Correction Map Synthesizing using Deep Neural Networks
title_full Towards a Whole Body [<sup>18</sup><i>F</i>] FDG Positron Emission Tomography Attenuation Correction Map Synthesizing using Deep Neural Networks
title_fullStr Towards a Whole Body [<sup>18</sup><i>F</i>] FDG Positron Emission Tomography Attenuation Correction Map Synthesizing using Deep Neural Networks
title_full_unstemmed Towards a Whole Body [<sup>18</sup><i>F</i>] FDG Positron Emission Tomography Attenuation Correction Map Synthesizing using Deep Neural Networks
title_sort towards a whole body [<sup>18</sup><i>f</i>] fdg positron emission tomography attenuation correction map synthesizing using deep neural networks
publishDate 2021
url http://sedici.unlp.edu.ar/handle/10915/118045
https://journal.info.unlp.edu.ar/JCST/article/view/1519
work_keys_str_mv AT rodriguezcolmeiroramiro towardsawholebodysup18supififdgpositronemissiontomographyattenuationcorrectionmapsynthesizingusingdeepneuralnetworks
AT verrastroclaudio towardsawholebodysup18supififdgpositronemissiontomographyattenuationcorrectionmapsynthesizingusingdeepneuralnetworks
AT minskydaniel towardsawholebodysup18supififdgpositronemissiontomographyattenuationcorrectionmapsynthesizingusingdeepneuralnetworks
AT grosgesthomas towardsawholebodysup18supififdgpositronemissiontomographyattenuationcorrectionmapsynthesizingusingdeepneuralnetworks
AT rodriguezcolmeiroramiro hacialasintetizaciondemapasdeatenuaciondecuerpocompletoparatomografiaporemisiondepositronesdesup18supififdgusandoredesneuronalesprofundas
AT verrastroclaudio hacialasintetizaciondemapasdeatenuaciondecuerpocompletoparatomografiaporemisiondepositronesdesup18supififdgusandoredesneuronalesprofundas
AT minskydaniel hacialasintetizaciondemapasdeatenuaciondecuerpocompletoparatomografiaporemisiondepositronesdesup18supififdgusandoredesneuronalesprofundas
AT grosgesthomas hacialasintetizaciondemapasdeatenuaciondecuerpocompletoparatomografiaporemisiondepositronesdesup18supififdgusandoredesneuronalesprofundas
bdutipo_str Repositorios
_version_ 1764820447346556931

Ejemplares similares