DeepTOFSino: A deep learning model for synthesizing full-dose time-of-flight bin sinograms from their corresponding low-dose sinograms

Purpose: Reducing the injected activity and/or the scanning time is a desirable goal to minimize radiation exposure and maximize patients’ comfort. To achieve this goal, we developed a deep neural network (DNN) model for synthesizing full-dose (FD) time-of-flight (TOF) bin sinograms from their corre...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Amirhossein Sanaat, Hossein Shooli, Sohrab Ferdowsi, Isaac Shiri, Hossein Arabi, Habib Zaidi
Formato:	article
Lenguaje:	EN
Publicado:	Elsevier 2021
Materias:	PET/CT Brain imaging Low-dose imaging Deep learning Time-of-flight Neurosciences. Biological psychiatry. Neuropsychiatry RC321-571
Acceso en línea:	https://doaj.org/article/6e6aa4f1cd8c46f89f61150479de2fdc
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:6e6aa4f1cd8c46f89f61150479de2fdc
record_format	dspace
spelling	oai:doaj.org-article:6e6aa4f1cd8c46f89f61150479de2fdc2021-11-10T04:21:13ZDeepTOFSino: A deep learning model for synthesizing full-dose time-of-flight bin sinograms from their corresponding low-dose sinograms1095-957210.1016/j.neuroimage.2021.118697https://doaj.org/article/6e6aa4f1cd8c46f89f61150479de2fdc2021-12-01T00:00:00Zhttp://www.sciencedirect.com/science/article/pii/S1053811921009708https://doaj.org/toc/1095-9572Purpose: Reducing the injected activity and/or the scanning time is a desirable goal to minimize radiation exposure and maximize patients’ comfort. To achieve this goal, we developed a deep neural network (DNN) model for synthesizing full-dose (FD) time-of-flight (TOF) bin sinograms from their corresponding fast/low-dose (LD) TOF bin sinograms. Methods: Clinical brain PET/CT raw data of 140 normal and abnormal patients were employed to create LD and FD TOF bin sinograms. The LD TOF sinograms were created through 5% undersampling of FD list-mode PET data. The TOF sinograms were split into seven time bins (0, ±1, ±2, ±3). Residual network (ResNet) algorithms were trained separately to generate FD bins from LD bins. An extra ResNet model was trained to synthesize FD images from LD images to compare the performance of DNN in sinogram space (SS) vs implementation in image space (IS). Comprehensive quantitative and statistical analysis was performed to assess the performance of the proposed model using established quantitative metrics, including the peak signal-to-noise ratio (PSNR), structural similarity index metric (SSIM) region-wise standardized uptake value (SUV) bias and statistical analysis for 83 brain regions. Results: SSIM and PSNR values of 0.97 ± 0.01, 0.98 ± 0.01 and 33.70 ± 0.32, 39.36 ± 0.21 were obtained for IS and SS, respectively, compared to 0.86 ± 0.02and 31.12 ± 0.22 for reference LD images. The absolute average SUV bias was 0.96 ± 0.95% and 1.40 ± 0.72% for SS and IS implementations, respectively. The joint histogram analysis revealed the lowest mean square error (MSE) and highest correlation (R2 = 0.99, MSE = 0.019) was achieved by SS compared to IS (R2 = 0.97, MSE= 0.028). The Bland & Altman analysis showed that the lowest SUV bias (-0.4%) and minimum variance (95% CI: -2.6%, +1.9%) were achieved by SS images. The voxel-wise t-test analysis revealed the presence of voxels with statistically significantly lower values in LD, IS, and SS images compared to FD images respectively. Conclusion: The results demonstrated that images reconstructed from the predicted TOF FD sinograms using the SS approach led to higher image quality and lower bias compared to images predicted from LD images.Amirhossein SanaatHossein ShooliSohrab FerdowsiIsaac ShiriHossein ArabiHabib ZaidiElsevierarticlePET/CTBrain imagingLow-dose imagingDeep learningTime-of-flightNeurosciences. Biological psychiatry. NeuropsychiatryRC321-571ENNeuroImage, Vol 245, Iss , Pp 118697- (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	PET/CT Brain imaging Low-dose imaging Deep learning Time-of-flight Neurosciences. Biological psychiatry. Neuropsychiatry RC321-571
spellingShingle	PET/CT Brain imaging Low-dose imaging Deep learning Time-of-flight Neurosciences. Biological psychiatry. Neuropsychiatry RC321-571 Amirhossein Sanaat Hossein Shooli Sohrab Ferdowsi Isaac Shiri Hossein Arabi Habib Zaidi DeepTOFSino: A deep learning model for synthesizing full-dose time-of-flight bin sinograms from their corresponding low-dose sinograms
description	Purpose: Reducing the injected activity and/or the scanning time is a desirable goal to minimize radiation exposure and maximize patients’ comfort. To achieve this goal, we developed a deep neural network (DNN) model for synthesizing full-dose (FD) time-of-flight (TOF) bin sinograms from their corresponding fast/low-dose (LD) TOF bin sinograms. Methods: Clinical brain PET/CT raw data of 140 normal and abnormal patients were employed to create LD and FD TOF bin sinograms. The LD TOF sinograms were created through 5% undersampling of FD list-mode PET data. The TOF sinograms were split into seven time bins (0, ±1, ±2, ±3). Residual network (ResNet) algorithms were trained separately to generate FD bins from LD bins. An extra ResNet model was trained to synthesize FD images from LD images to compare the performance of DNN in sinogram space (SS) vs implementation in image space (IS). Comprehensive quantitative and statistical analysis was performed to assess the performance of the proposed model using established quantitative metrics, including the peak signal-to-noise ratio (PSNR), structural similarity index metric (SSIM) region-wise standardized uptake value (SUV) bias and statistical analysis for 83 brain regions. Results: SSIM and PSNR values of 0.97 ± 0.01, 0.98 ± 0.01 and 33.70 ± 0.32, 39.36 ± 0.21 were obtained for IS and SS, respectively, compared to 0.86 ± 0.02and 31.12 ± 0.22 for reference LD images. The absolute average SUV bias was 0.96 ± 0.95% and 1.40 ± 0.72% for SS and IS implementations, respectively. The joint histogram analysis revealed the lowest mean square error (MSE) and highest correlation (R2 = 0.99, MSE = 0.019) was achieved by SS compared to IS (R2 = 0.97, MSE= 0.028). The Bland & Altman analysis showed that the lowest SUV bias (-0.4%) and minimum variance (95% CI: -2.6%, +1.9%) were achieved by SS images. The voxel-wise t-test analysis revealed the presence of voxels with statistically significantly lower values in LD, IS, and SS images compared to FD images respectively. Conclusion: The results demonstrated that images reconstructed from the predicted TOF FD sinograms using the SS approach led to higher image quality and lower bias compared to images predicted from LD images.
format	article
author	Amirhossein Sanaat Hossein Shooli Sohrab Ferdowsi Isaac Shiri Hossein Arabi Habib Zaidi
author_facet	Amirhossein Sanaat Hossein Shooli Sohrab Ferdowsi Isaac Shiri Hossein Arabi Habib Zaidi
author_sort	Amirhossein Sanaat
title	DeepTOFSino: A deep learning model for synthesizing full-dose time-of-flight bin sinograms from their corresponding low-dose sinograms
title_short	DeepTOFSino: A deep learning model for synthesizing full-dose time-of-flight bin sinograms from their corresponding low-dose sinograms
title_full	DeepTOFSino: A deep learning model for synthesizing full-dose time-of-flight bin sinograms from their corresponding low-dose sinograms
title_fullStr	DeepTOFSino: A deep learning model for synthesizing full-dose time-of-flight bin sinograms from their corresponding low-dose sinograms
title_full_unstemmed	DeepTOFSino: A deep learning model for synthesizing full-dose time-of-flight bin sinograms from their corresponding low-dose sinograms
title_sort	deeptofsino: a deep learning model for synthesizing full-dose time-of-flight bin sinograms from their corresponding low-dose sinograms
publisher	Elsevier
publishDate	2021
url	https://doaj.org/article/6e6aa4f1cd8c46f89f61150479de2fdc
work_keys_str_mv	AT amirhosseinsanaat deeptofsinoadeeplearningmodelforsynthesizingfulldosetimeofflightbinsinogramsfromtheircorrespondinglowdosesinograms AT hosseinshooli deeptofsinoadeeplearningmodelforsynthesizingfulldosetimeofflightbinsinogramsfromtheircorrespondinglowdosesinograms AT sohrabferdowsi deeptofsinoadeeplearningmodelforsynthesizingfulldosetimeofflightbinsinogramsfromtheircorrespondinglowdosesinograms AT isaacshiri deeptofsinoadeeplearningmodelforsynthesizingfulldosetimeofflightbinsinogramsfromtheircorrespondinglowdosesinograms AT hosseinarabi deeptofsinoadeeplearningmodelforsynthesizingfulldosetimeofflightbinsinogramsfromtheircorrespondinglowdosesinograms AT habibzaidi deeptofsinoadeeplearningmodelforsynthesizingfulldosetimeofflightbinsinogramsfromtheircorrespondinglowdosesinograms
_version_	1718440659019366400

DeepTOFSino: A deep learning model for synthesizing full-dose time-of-flight bin sinograms from their corresponding low-dose sinograms

Ejemplares similares