Progressively volumetrized deep generative models for data-efficient contextual learning of MR image recovery

Simple item page
buir.contributor.authorYurt, Mahmut
buir.contributor.authorÖzbey, Muzaffer
buir.contributor.authorDar, Salman U.H.
buir.contributor.authorTınaz, Berk
buir.contributor.authorOğuz, Kader K.
buir.contributor.authorÇukur, Tolga
buir.contributor.orcidÇukur, Tolga|0000-0002-2296-851X
dc.citation.epage[19]en_US
dc.citation.spage[1]en_US
dc.citation.volumeNumber78en_US
dc.contributor.authorYurt, Mahmut
dc.contributor.authorÖzbey, Muzaffer
dc.contributor.authorDar, Salman U.H.
dc.contributor.authorTınaz, Berk
dc.contributor.authorOğuz, Kader K.
dc.contributor.authorÇukur, Tolga
dc.coverage.spatialNetherlandsen_US
dc.date.accessioned2023-02-21T13:44:51Z
dc.date.available2023-02-21T13:44:51Z
dc.date.issued2022-05
dc.departmentDepartment of Electrical and Electronics Engineeringen_US
dc.departmentInterdisciplinary Program in Neuroscience (NEUROSCIENCE)en_US
dc.departmentNational Magnetic Resonance Research Center (UMRAM)en_US
dc.description.abstractMagnetic resonance imaging (MRI) offers the flexibility to image a given anatomic volume under a multi- tude of tissue contrasts. Yet, scan time considerations put stringent limits on the quality and diversity of MRI data. The gold-standard approach to alleviate this limitation is to recover high-quality images from data undersampled across various dimensions, most commonly the Fourier domain or contrast sets. A primary distinction among recovery methods is whether the anatomy is processed per volume or per cross-section. Volumetric models offer enhanced capture of global contextual information, but they can suffer from suboptimal learning due to elevated model complexity. Cross-sectional models with lower complexity offer improved learning behavior, yet they ignore contextual information across the longitu- dinal dimension of the volume. Here, we introduce a novel progressive volumetrization strategy for gen- erative models (ProvoGAN) that serially decomposes complex volumetric image recovery tasks into suc- cessive cross-sectional mappings task-optimally ordered across individual rectilinear dimensions. Provo-GAN effectively captures global context and recovers fine-structural details across all dimensions, while maintaining low model complexity and improved learning behavior. Comprehensive demonstrations on mainstream MRI reconstruction and synthesis tasks show that ProvoGAN yields superior performance to state-of-the-art volumetric and cross-sectional models.en_US
dc.description.provenanceSubmitted by Betül Özen (ozen@bilkent.edu.tr) on 2023-02-21T13:44:51Z No. of bitstreams: 1 Progressively_volumetrized_deep_generative_models_for_data-efficient_contextual_learning_of_MR_image_recovery.pdf: 4622550 bytes, checksum: 229c83f1456b9420095db5d62687aa81 (MD5)en
dc.description.provenanceMade available in DSpace on 2023-02-21T13:44:51Z (GMT). No. of bitstreams: 1 Progressively_volumetrized_deep_generative_models_for_data-efficient_contextual_learning_of_MR_image_recovery.pdf: 4622550 bytes, checksum: 229c83f1456b9420095db5d62687aa81 (MD5) Previous issue date: 2022-05en
dc.identifier.doi10.1016/j.media.2022.102429en_US
dc.identifier.eissn1361-8423
dc.identifier.issn1361-8415
dc.identifier.urihttp://hdl.handle.net/11693/111593
dc.language.isoEnglishen_US
dc.publisherElsevier BVen_US
dc.relation.isversionofhttps://doi.org/10.1016/j.media.2022.102429en_US
dc.source.titleMedical Image Analysisen_US
dc.subjectMRIen_US
dc.subjectGenerative adversarial networksen_US
dc.subjectSynthesisen_US
dc.subjectReconstructionen_US
dc.subjectCross-sectionen_US
dc.subjectModel complexityen_US
dc.subjectContexten_US
dc.titleProgressively volumetrized deep generative models for data-efficient contextual learning of MR image recoveryen_US
dc.typeArticleen_US

Files

Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
Progressively_volumetrized_deep_generative_models_for_data-efficient_contextual_learning_of_MR_image_recovery.pdf
Size:
4.41 MB
Format:
Adobe Portable Document Format
Description:
Download
License bundle
Now showing 1 - 1 of 1
No Thumbnail Available
Name:
license.txt
Size:
1.69 KB
Format:
Item-specific license agreed upon to submission
Description:
Download

Collections

Scholarly Publications - Electrical and Electronics Engineering