Generating 3D TOF-MRA volumes and segmentation labels using generative adversarial networks

dc.contributor.authorSubramaniam, Pooja
dc.contributor.authorKossen, Tabea
dc.contributor.authorRitter, Kerstin
dc.contributor.authorHennemuth, Anja
dc.contributor.authorHildebrand, Kristian
dc.contributor.authorHilbert, Adam
dc.contributor.authorSobesky, Jan
dc.contributor.authorLivne, Michelle
dc.contributor.authorGalinovic, Ivana
dc.contributor.authorKhalil, Ahmed A.
dc.contributor.authorFiebach, Jochen B.
dc.contributor.authorFrey, Dietmar
dc.contributor.authorMadai, Vince I.
dc.date.accessioned2022-12-29T12:35:39Z
dc.date.available2022-12-29T12:35:39Z
dc.date.issued2022-02-24
dc.description.abstractDeep learning requires large labeled datasets that are difficult to gather in medical imaging due to data privacy issues and time-consuming manual labeling. Generative Adversarial Networks (GANs) can alleviate these challenges enabling synthesis of shareable data. While 2D GANs have been used to generate 2D images with their corresponding labels, they cannot capture the volumetric information of 3D medical imaging. 3D GANs are more suitable for this and have been used to generate 3D volumes but not their corresponding labels. One reason might be that synthesizing 3D volumes is challenging owing to computational limitations. In this work, we present 3D GANs for the generation of 3D medical image volumes with corresponding labels applying mixed precision to alleviate computational constraints. We generated 3D Time-of-Flight Magnetic Resonance Angiography (TOF-MRA) patches with their corresponding brain blood vessel segmentation labels. We used four variants of 3D Wasserstein GAN (WGAN) with: 1) gradient penalty (GP), 2) GP with spectral normalization (SN), 3) SN with mixed precision (SN-MP), and 4) SN-MP with double filters per layer (c-SN-MP). The generated patches were quantitatively evaluated using the Fréchet Inception Distance (FID) and Precision and Recall of Distributions (PRD). Further, 3D U-Nets were trained with patch-label pairs from different WGAN models and their performance was compared to the performance of a benchmark U-Net trained on real data. The segmentation performance of all U-Net models was assessed using Dice Similarity Coefficient (DSC) and balanced Average Hausdorff Distance (bAVD) for a) all vessels, and b) intracranial vessels only. Our results show that patches generated with WGAN models using mixed precision (SN-MP and c-SN-MP) yielded the lowest FID scores and the best PRD curves. Among the 3D U-Nets trained with synthetic patch-label pairs, c-SN-MP pairs achieved the highest DSC (0.841) and lowest bAVD (0.508) compared to the benchmark U-Net trained on real data (DSC 0.901; bAVD 0.294) for intracranial vessels. In conclusion, our solution generates realistic 3D TOF-MRA patches and labels for brain vessel segmentation. We demonstrate the benefit of using mixed precision for computational efficiency resulting in the best-performing GAN-architecture. Our work paves the way towards sharing of labeled 3D medical data which would increase generalizability of deep learning models for clinical use.en
dc.identifier.eissn1361-8423
dc.identifier.issn1361-8415
dc.identifier.urihttps://depositonce.tu-berlin.de/handle/11303/17917
dc.identifier.urihttps://doi.org/10.14279/depositonce-16706
dc.language.isoen
dc.relation.ispartof10.14279/depositonce-16401
dc.rights.urihttps://creativecommons.org/licenses/by/4.0/
dc.subject.ddc610 Medizin und Gesundheitde
dc.subject.ddc006 Spezielle Computerverfahrende
dc.subject.othergenerative adversarial networksen
dc.subject.other3D medical imagingen
dc.subject.othermixed precisionen
dc.subject.otheranonymizationen
dc.subject.otherbrain vessel segmentationen
dc.titleGenerating 3D TOF-MRA volumes and segmentation labels using generative adversarial networksen
dc.typeArticle
dc.type.versionpublishedVersion
dcterms.bibliographicCitation.articlenumber102396
dcterms.bibliographicCitation.doi10.1016/j.media.2022.102396
dcterms.bibliographicCitation.journaltitleMedical Image Analysis
dcterms.bibliographicCitation.originalpublishernameElsevier
dcterms.bibliographicCitation.originalpublisherplaceAmsterdam
dcterms.bibliographicCitation.volume78
dcterms.rightsHolder.referenceCreative-Commons-Lizenz
tub.accessrights.dnbfree
tub.affiliationFak. 4 Elektrotechnik und Informatik::Inst. Technische Informatik und Mikroelektronik::FG Computer Vision & Remote Sensing
tub.publisher.universityorinstitutionTechnische Universität Berlin

Files

Original bundle
Now showing 1 - 1 of 1
Loading…
Thumbnail Image
Name:
Subramaniam_etal_Generating_2022.pdf
Size:
3.37 MB
Format:
Adobe Portable Document Format
License bundle
Now showing 1 - 1 of 1
No Thumbnail Available
Name:
license.txt
Size:
4.23 KB
Format:
Item-specific license agreed upon to submission
Description:

Collections