Cargando…

Image-to-image generative adversarial networks for synthesizing perfusion parameter maps from DSC-MR images in cerebrovascular disease

Stroke is a major cause of death or disability. As imaging-based patient stratification improves acute stroke therapy, dynamic susceptibility contrast magnetic resonance imaging (DSC-MRI) is of major interest in image brain perfusion. However, expert-level perfusion maps require a manual or semi-man...

Descripción completa

Detalles Bibliográficos
Autores principales: Kossen, Tabea, Madai, Vince I., Mutke, Matthias A., Hennemuth, Anja, Hildebrand, Kristian, Behland, Jonas, Aslan, Cagdas, Hilbert, Adam, Sobesky, Jan, Bendszus, Martin, Frey, Dietmar
Formato: Online Artículo Texto
Lenguaje:English
Publicado: Frontiers Media S.A. 2023
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9871486/
https://www.ncbi.nlm.nih.gov/pubmed/36703627
http://dx.doi.org/10.3389/fneur.2022.1051397
_version_ 1784877184263913472
author Kossen, Tabea
Madai, Vince I.
Mutke, Matthias A.
Hennemuth, Anja
Hildebrand, Kristian
Behland, Jonas
Aslan, Cagdas
Hilbert, Adam
Sobesky, Jan
Bendszus, Martin
Frey, Dietmar
author_facet Kossen, Tabea
Madai, Vince I.
Mutke, Matthias A.
Hennemuth, Anja
Hildebrand, Kristian
Behland, Jonas
Aslan, Cagdas
Hilbert, Adam
Sobesky, Jan
Bendszus, Martin
Frey, Dietmar
author_sort Kossen, Tabea
collection PubMed
description Stroke is a major cause of death or disability. As imaging-based patient stratification improves acute stroke therapy, dynamic susceptibility contrast magnetic resonance imaging (DSC-MRI) is of major interest in image brain perfusion. However, expert-level perfusion maps require a manual or semi-manual post-processing by a medical expert making the procedure time-consuming and less-standardized. Modern machine learning methods such as generative adversarial networks (GANs) have the potential to automate the perfusion map generation on an expert level without manual validation. We propose a modified pix2pix GAN with a temporal component (temp-pix2pix-GAN) that generates perfusion maps in an end-to-end fashion. We train our model on perfusion maps infused with expert knowledge to encode it into the GANs. The performance was trained and evaluated using the structural similarity index measure (SSIM) on two datasets including patients with acute stroke and the steno-occlusive disease. Our temp-pix2pix architecture showed high performance on the acute stroke dataset for all perfusion maps (mean SSIM 0.92–0.99) and good performance on data including patients with the steno-occlusive disease (mean SSIM 0.84–0.99). While clinical validation is still necessary for future studies, our results mark an important step toward automated expert-level perfusion maps and thus fast patient stratification.
format Online
Article
Text
id pubmed-9871486
institution National Center for Biotechnology Information
language English
publishDate 2023
publisher Frontiers Media S.A.
record_format MEDLINE/PubMed
spelling pubmed-98714862023-01-25 Image-to-image generative adversarial networks for synthesizing perfusion parameter maps from DSC-MR images in cerebrovascular disease Kossen, Tabea Madai, Vince I. Mutke, Matthias A. Hennemuth, Anja Hildebrand, Kristian Behland, Jonas Aslan, Cagdas Hilbert, Adam Sobesky, Jan Bendszus, Martin Frey, Dietmar Front Neurol Neurology Stroke is a major cause of death or disability. As imaging-based patient stratification improves acute stroke therapy, dynamic susceptibility contrast magnetic resonance imaging (DSC-MRI) is of major interest in image brain perfusion. However, expert-level perfusion maps require a manual or semi-manual post-processing by a medical expert making the procedure time-consuming and less-standardized. Modern machine learning methods such as generative adversarial networks (GANs) have the potential to automate the perfusion map generation on an expert level without manual validation. We propose a modified pix2pix GAN with a temporal component (temp-pix2pix-GAN) that generates perfusion maps in an end-to-end fashion. We train our model on perfusion maps infused with expert knowledge to encode it into the GANs. The performance was trained and evaluated using the structural similarity index measure (SSIM) on two datasets including patients with acute stroke and the steno-occlusive disease. Our temp-pix2pix architecture showed high performance on the acute stroke dataset for all perfusion maps (mean SSIM 0.92–0.99) and good performance on data including patients with the steno-occlusive disease (mean SSIM 0.84–0.99). While clinical validation is still necessary for future studies, our results mark an important step toward automated expert-level perfusion maps and thus fast patient stratification. Frontiers Media S.A. 2023-01-10 /pmc/articles/PMC9871486/ /pubmed/36703627 http://dx.doi.org/10.3389/fneur.2022.1051397 Text en Copyright © 2023 Kossen, Madai, Mutke, Hennemuth, Hildebrand, Behland, Aslan, Hilbert, Sobesky, Bendszus and Frey. https://creativecommons.org/licenses/by/4.0/This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
spellingShingle Neurology
Kossen, Tabea
Madai, Vince I.
Mutke, Matthias A.
Hennemuth, Anja
Hildebrand, Kristian
Behland, Jonas
Aslan, Cagdas
Hilbert, Adam
Sobesky, Jan
Bendszus, Martin
Frey, Dietmar
Image-to-image generative adversarial networks for synthesizing perfusion parameter maps from DSC-MR images in cerebrovascular disease
title Image-to-image generative adversarial networks for synthesizing perfusion parameter maps from DSC-MR images in cerebrovascular disease
title_full Image-to-image generative adversarial networks for synthesizing perfusion parameter maps from DSC-MR images in cerebrovascular disease
title_fullStr Image-to-image generative adversarial networks for synthesizing perfusion parameter maps from DSC-MR images in cerebrovascular disease
title_full_unstemmed Image-to-image generative adversarial networks for synthesizing perfusion parameter maps from DSC-MR images in cerebrovascular disease
title_short Image-to-image generative adversarial networks for synthesizing perfusion parameter maps from DSC-MR images in cerebrovascular disease
title_sort image-to-image generative adversarial networks for synthesizing perfusion parameter maps from dsc-mr images in cerebrovascular disease
topic Neurology
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9871486/
https://www.ncbi.nlm.nih.gov/pubmed/36703627
http://dx.doi.org/10.3389/fneur.2022.1051397
work_keys_str_mv AT kossentabea imagetoimagegenerativeadversarialnetworksforsynthesizingperfusionparametermapsfromdscmrimagesincerebrovasculardisease
AT madaivincei imagetoimagegenerativeadversarialnetworksforsynthesizingperfusionparametermapsfromdscmrimagesincerebrovasculardisease
AT mutkematthiasa imagetoimagegenerativeadversarialnetworksforsynthesizingperfusionparametermapsfromdscmrimagesincerebrovasculardisease
AT hennemuthanja imagetoimagegenerativeadversarialnetworksforsynthesizingperfusionparametermapsfromdscmrimagesincerebrovasculardisease
AT hildebrandkristian imagetoimagegenerativeadversarialnetworksforsynthesizingperfusionparametermapsfromdscmrimagesincerebrovasculardisease
AT behlandjonas imagetoimagegenerativeadversarialnetworksforsynthesizingperfusionparametermapsfromdscmrimagesincerebrovasculardisease
AT aslancagdas imagetoimagegenerativeadversarialnetworksforsynthesizingperfusionparametermapsfromdscmrimagesincerebrovasculardisease
AT hilbertadam imagetoimagegenerativeadversarialnetworksforsynthesizingperfusionparametermapsfromdscmrimagesincerebrovasculardisease
AT sobeskyjan imagetoimagegenerativeadversarialnetworksforsynthesizingperfusionparametermapsfromdscmrimagesincerebrovasculardisease
AT bendszusmartin imagetoimagegenerativeadversarialnetworksforsynthesizingperfusionparametermapsfromdscmrimagesincerebrovasculardisease
AT freydietmar imagetoimagegenerativeadversarialnetworksforsynthesizingperfusionparametermapsfromdscmrimagesincerebrovasculardisease