Faithful Synthesis of Low-Dose Contrast-Enhanced Brain MRI Scans Using Noise-Preserving Conditional GANs

Greenspan, Hayit; Taylor, Russell; Madabhushi, Anant; Deike-Hofmann, Katerina; Syeda-Mahmood, Tanveer; Radbruch, Alexander; Pinetz, Thomas; Mousavi, Parvin; Duncan, James; Kobler, Erich; Salcudean, Septimiu; Effland, Alexander; Haase, Robert

doi:10.1007/978-3-031-43895-0_57

Contribution to a conference proceedings/Contribution to a book

DZNE-2023-01039

Faithful Synthesis of Low-Dose Contrast-Enhanced Brain MRI Scans Using Noise-Preserving Conditional GANs

Greenspan, H. (Editor) ; Madabhushi, A. (Editor) ; Mousavi, P. (Editor) ; Salcudean, S. (Editor) ; Duncan, J. (Editor) ; Syeda-Mahmood, T. (Editor) ; Taylor, R. (Editor) ; Pinetz, T. ; Kobler, E. ; Haase, R. ; Deike-Hofmann, K.DZNE* ; Radbruch, A.DZNE* ; Effland, A.Extern*

2023
Springer Nature Switzerland Cham
ISBN: 978-3-031-43894-3 (print), 978-3-031-43895-0 (electronic)

Medical Image Computing and Computer Assisted Intervention – MICCAI 2023 / Greenspan, Hayit (Editor) ; Cham : Springer Nature Switzerland, 2023, Chapter 57 ; ISSN: 0302-9743=1611-3349 ; ISBN: 978-3-031-43894-3=978-3-031-43895-0 ; doi:10.1007/978-3-031-43895-0
26th International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2023, Vancouver, Canada, 8 Oct 2023 - 12 Oct 2023 Cham : Springer Nature Switzerland, Lecture Notes in Computer Science 14221, 607 - 617 (2023) [10.1007/978-3-031-43895-0_57]

This record in other databases:

Please use a persistent id in citations: doi:10.1007/978-3-031-43895-0_57

Abstract: Today Gadolinium-based contrast agents (GBCA) are indispensable in Magnetic Resonance Imaging (MRI) for diagnosing various diseases. However, GBCAs are expensive and may accumulate in patients with potential side effects, thus dose-reduction is recommended. Still, it is unclear to which extent the GBCA dose can be reduced while preserving the diagnostic value – especially in pathological regions. To address this issue, we collected brain MRI scans at numerous non-standard GBCA dosages and developed a conditional GAN model for synthesizing corresponding images at fractional dose levels. Along with the adversarial loss, we advocate a novel content loss function based on the Wasserstein distance of locally paired patch statistics for the faithful preservation of noise. Our numerical experiments show that conditional GANs are suitable for generating images at different GBCA dose levels and can be used to augment datasets for virtual contrast models. Moreover, our model can be transferred to openly available datasets such as BraTS, where non-standard GBCA dosage images do not exist. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023.

Contributing Institute(s):