Federated learning of generative ımage priors for MRI reconstruction

Simple item page
buir.contributor.authorElmas, Gökberk
buir.contributor.authorDar, Salman UH.
buir.contributor.authorKorkmaz, Yilmaz
buir.contributor.authorSusam, Burak
buir.contributor.authorOzbey, Muzaffer
buir.contributor.authorÇukur, Tolga
buir.contributor.orcidÇukur, Tolga|0000-0002-2296-851X
buir.contributor.orcidElmas, Gökberk|0000-0003-0124-6048
dc.citation.epage13en_US
dc.citation.spage1en_US
dc.contributor.authorElmas, Gökberk
dc.contributor.authorDar, Salman UH.
dc.contributor.authorKorkmaz, Yilmaz
dc.contributor.authorCeyani, E.
dc.contributor.authorSusam, Burak
dc.contributor.authorOzbey, Muzaffer
dc.contributor.authorAvestimehr, S.
dc.contributor.authorÇukur, Tolga
dc.date.accessioned2023-02-16T07:15:48Z
dc.date.available2023-02-16T07:15:48Z
dc.date.issued2022-11-09
dc.departmentDepartment of Electrical and Electronics Engineeringen_US
dc.description.abstractMulti-institutional efforts can facilitate training of deep MRI reconstruction models, albeit privacy risks arise during cross-site sharing of imaging data. Federated learning (FL) has recently been introduced to address privacy concerns by enabling distributed training without transfer of imaging data. Existing FL methods employ conditional reconstruction models to map from undersampled to fully-sampled acquisitions via explicit knowledge of the accelerated imaging operator. Since conditional models generalize poorly across different acceleration rates or sampling densities, imaging operators must be fixed between training and testing, and they are typically matched across sites. To improve patient privacy, performance and flexibility in multi-site collaborations, here we introduce Federated learning of Generative IMage Priors (FedGIMP) for MRI reconstruction. FedGIMP leverages a two-stage approach: cross-site learning of a generative MRI prior, and prior adaptation following injection of the imaging operator. The global MRI prior is learned via an unconditional adversarial model that synthesizes high-quality MR images based on latent variables. A novel mapper subnetwork produces site-specific latents to maintain specificity in the prior. During inference, the prior is first combined with subject-specific imaging operators to enable reconstruction, and it is then adapted to individual cross-sections by minimizing a data-consistency loss. Comprehensive experiments on multi-institutional datasets clearly demonstrate enhanced performance of FedGIMP against both centralized and FL methods based on conditional modelsen_US
dc.identifier.doi10.1109/TMI.2022.3220757en_US
dc.identifier.eissn1558-254X
dc.identifier.issn0278-0062
dc.identifier.urihttp://hdl.handle.net/11693/111391
dc.language.isoEnglishen_US
dc.publisherInstitute of Electrical and Electronics Engineers Inc.en_US
dc.relation.isversionofhttps://www.doi.org/10.1109/TMI.2022.3220757en_US
dc.source.titleIEEE Transactions on Medical Imagingen_US
dc.subjectMRIen_US
dc.subjectAccelerateden_US
dc.subjectReconstructionen_US
dc.subjectGenerativeen_US
dc.subjectPrioren_US
dc.subjectFederated learningen_US
dc.subjectDistributeden_US
dc.subjectCollaborativeen_US
dc.titleFederated learning of generative ımage priors for MRI reconstructionen_US
dc.typeArticleen_US

Files

Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
Federated_learning_of_generative_ımage_priors_for_MRI_reconstruction.pdf
Size:
9.09 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

Department of Electrical and Electronics Engineering