Forskning
Udskriv Udskriv
Switch language
Region Hovedstaden - en del af Københavns Universitetshospital
Udgivet

Deep learning based low-activity PET reconstruction of [11C]PiB and [18F]FE-PE2I in neurodegenerative disorders

Publikation: Bidrag til tidsskriftTidsskriftartikelpeer review

Harvard

APA

CBE

MLA

Vancouver

Author

Bibtex

@article{95d131c89dd0445f8236b4da171b52a3,
title = "Deep learning based low-activity PET reconstruction of [11C]PiB and [18F]FE-PE2I in neurodegenerative disorders",
abstract = "PURPOSE: Positron Emission Tomography (PET) can support a diagnosis of neurodegenerative disorder by identifying disease-specific pathologies. Our aim was to investigate the feasibility of using activity reduction in clinical [18F]FE-PE2I and [11C]PiB PET/CT scans, simulating low injected activity or scanning time reduction, in combination with AI-assisted denoising.METHODS: A total of 162 patients with clinically uncertain Alzheimer's disease underwent amyloid [11C]PiB PET/CT and 509 patients referred for clinically uncertain Parkinson's disease underwent dopamine transporter (DAT) [18F]FE-PE2I PET/CT. Simulated low-activity data were obtained by random sampling of 5% of the events from the list-mode file and a 5% time window extraction in the middle of the scan. A three-dimensional convolutional neural network (CNN) was trained to denoise the resulting PET images for each disease cohort.RESULTS: Noise reduction of low-activity PET images was successful for both cohorts using 5% of the original activity with improvement in visual quality and all similarity metrics with respect to the ground-truth images. Clinically relevant metrics extracted from the low-activity images deviated <2% compared to ground-truth values, which were not significantly changed when extracting the metrics from the denoised images.CONCLUSION: The presented models were based on the same network architecture and proved to be a robust tool for denoising brain PET images with two widely different tracer distributions (delocalized, ([11C]PiB, and highly localized, [18F]FE-PE2I). This broad and robust application makes the presented network a good choice for improving the quality of brain images to the level of the standard-activity images without degrading clinical metric extraction. This will allow for reduced dose or scan time in PET/CT to be implemented clinically.",
keywords = "Deep Learning, Humans, Nortropanes, Parkinson Disease, Positron Emission Tomography Computed Tomography, Positron-Emission Tomography/methods, Deep learning, [ F]FE-PE2I, Parkinson's disease, PET denoising, [ C]PiB, Alzheimer's disease",
author = "Daveau, {Rapha{\"e}l S} and Ian Law and Henriksen, {Otto M{\o}lby} and Hasselbalch, {Steen Gregers} and Andersen, {Ulrik Bj{\o}rn} and Lasse Anderberg and Liselotte H{\o}jgaard and Andersen, {Flemming Littrup} and Ladefoged, {Claes N{\o}hr}",
note = "Copyright {\textcopyright} 2022. Published by Elsevier Inc.",
year = "2022",
month = oct,
day = "1",
doi = "10.1016/j.neuroimage.2022.119412",
language = "English",
volume = "259",
pages = "119412",
journal = "NeuroImage",
issn = "1053-8119",
publisher = "Academic Press",

}

RIS

TY - JOUR

T1 - Deep learning based low-activity PET reconstruction of [11C]PiB and [18F]FE-PE2I in neurodegenerative disorders

AU - Daveau, Raphaël S

AU - Law, Ian

AU - Henriksen, Otto Mølby

AU - Hasselbalch, Steen Gregers

AU - Andersen, Ulrik Bjørn

AU - Anderberg, Lasse

AU - Højgaard, Liselotte

AU - Andersen, Flemming Littrup

AU - Ladefoged, Claes Nøhr

N1 - Copyright © 2022. Published by Elsevier Inc.

PY - 2022/10/1

Y1 - 2022/10/1

N2 - PURPOSE: Positron Emission Tomography (PET) can support a diagnosis of neurodegenerative disorder by identifying disease-specific pathologies. Our aim was to investigate the feasibility of using activity reduction in clinical [18F]FE-PE2I and [11C]PiB PET/CT scans, simulating low injected activity or scanning time reduction, in combination with AI-assisted denoising.METHODS: A total of 162 patients with clinically uncertain Alzheimer's disease underwent amyloid [11C]PiB PET/CT and 509 patients referred for clinically uncertain Parkinson's disease underwent dopamine transporter (DAT) [18F]FE-PE2I PET/CT. Simulated low-activity data were obtained by random sampling of 5% of the events from the list-mode file and a 5% time window extraction in the middle of the scan. A three-dimensional convolutional neural network (CNN) was trained to denoise the resulting PET images for each disease cohort.RESULTS: Noise reduction of low-activity PET images was successful for both cohorts using 5% of the original activity with improvement in visual quality and all similarity metrics with respect to the ground-truth images. Clinically relevant metrics extracted from the low-activity images deviated <2% compared to ground-truth values, which were not significantly changed when extracting the metrics from the denoised images.CONCLUSION: The presented models were based on the same network architecture and proved to be a robust tool for denoising brain PET images with two widely different tracer distributions (delocalized, ([11C]PiB, and highly localized, [18F]FE-PE2I). This broad and robust application makes the presented network a good choice for improving the quality of brain images to the level of the standard-activity images without degrading clinical metric extraction. This will allow for reduced dose or scan time in PET/CT to be implemented clinically.

AB - PURPOSE: Positron Emission Tomography (PET) can support a diagnosis of neurodegenerative disorder by identifying disease-specific pathologies. Our aim was to investigate the feasibility of using activity reduction in clinical [18F]FE-PE2I and [11C]PiB PET/CT scans, simulating low injected activity or scanning time reduction, in combination with AI-assisted denoising.METHODS: A total of 162 patients with clinically uncertain Alzheimer's disease underwent amyloid [11C]PiB PET/CT and 509 patients referred for clinically uncertain Parkinson's disease underwent dopamine transporter (DAT) [18F]FE-PE2I PET/CT. Simulated low-activity data were obtained by random sampling of 5% of the events from the list-mode file and a 5% time window extraction in the middle of the scan. A three-dimensional convolutional neural network (CNN) was trained to denoise the resulting PET images for each disease cohort.RESULTS: Noise reduction of low-activity PET images was successful for both cohorts using 5% of the original activity with improvement in visual quality and all similarity metrics with respect to the ground-truth images. Clinically relevant metrics extracted from the low-activity images deviated <2% compared to ground-truth values, which were not significantly changed when extracting the metrics from the denoised images.CONCLUSION: The presented models were based on the same network architecture and proved to be a robust tool for denoising brain PET images with two widely different tracer distributions (delocalized, ([11C]PiB, and highly localized, [18F]FE-PE2I). This broad and robust application makes the presented network a good choice for improving the quality of brain images to the level of the standard-activity images without degrading clinical metric extraction. This will allow for reduced dose or scan time in PET/CT to be implemented clinically.

KW - Deep Learning

KW - Humans

KW - Nortropanes

KW - Parkinson Disease

KW - Positron Emission Tomography Computed Tomography

KW - Positron-Emission Tomography/methods

KW - Deep learning

KW - [ F]FE-PE2I

KW - Parkinson's disease

KW - PET denoising

KW - [ C]PiB

KW - Alzheimer's disease

UR - http://www.scopus.com/inward/record.url?scp=85133215660&partnerID=8YFLogxK

U2 - 10.1016/j.neuroimage.2022.119412

DO - 10.1016/j.neuroimage.2022.119412

M3 - Journal article

C2 - 35753592

VL - 259

SP - 119412

JO - NeuroImage

JF - NeuroImage

SN - 1053-8119

M1 - 119412

ER -

ID: 79048744