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

• Verfasst von: Kossen, Tabea [VerfasserIn]
• Verfasst von: Madai, Vince Istvan [VerfasserIn]
• Verfasst von: Mutke, Matthias Anthony [VerfasserIn]
• Verfasst von: Hennemuth, Anja [VerfasserIn]
• Verfasst von: Hildebrand, Kristian [VerfasserIn]
• Verfasst von: Behland, Jonas [VerfasserIn]
• Verfasst von: Aslan, Cagdas [VerfasserIn]
• Verfasst von: Hilbert, Adam [VerfasserIn]
• Verfasst von: Sobesky, Jan [VerfasserIn]
• Verfasst von: Bendszus, Martin [VerfasserIn]
• Verfasst von: Frey, Dietmar [VerfasserIn]
• Titel: Image-to-image generative adversarial networks for synthesizing perfusion parameter maps from DSC-MR images in cerebrovascular disease
• Verf.angabe: Tabea Kossen, Vince I. Madai, Matthias A. Mutke, Anja Hennemuth, Kristian Hildebrand, Jonas Behland, Cagdas Aslan, Adam Hilbert, Jan Sobesky, Martin Bendszus and Dietmar Frey
• E-Jahr: 2023
• Jahr: 10 January 2023
• Umfang: 14 S.
• Fussnoten: Gesehen am 13.06.2023
• Titel Quelle: Enthalten in: Frontiers in neurology
• Ort Quelle: Lausanne : Frontiers Research Foundation, 2008
• Jahr Quelle: 2023
• Band/Heft Quelle: 13(2023) vom: Jan., Artikel-ID 1051397, Seite 1-14
• ISSN Quelle: 1664-2295
• DOI: doi:10.3389/fneur.2022.1051397
• Datenträger: Online-Ressource
• Sprache: eng
• K10plus-PPN: 184888818X

Abstract

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.