000286250 001__ 286250
000286250 005__ 20240229155124.0
000286250 0247_ $$2doi$$a10.1016/j.media.2023.103059
000286250 0247_ $$2pmid$$apmid:38104402
000286250 0247_ $$2ISSN$$a1361-8415
000286250 0247_ $$2ISSN$$a1361-8431
000286250 0247_ $$2ISSN$$a1361-8423
000286250 0247_ $$2altmetric$$aaltmetric:157519096
000286250 037__ $$aDKFZ-2023-02741
000286250 041__ $$aEnglish
000286250 082__ $$a610
000286250 1001_ $$aTruhn, Daniel$$b0
000286250 245__ $$aEncrypted federated learning for secure decentralized collaboration in cancer image analysis.
000286250 260__ $$aAmsterdam [u.a.]$$bElsevier Science$$c2023
000286250 3367_ $$2DRIVER$$aarticle
000286250 3367_ $$2DataCite$$aOutput Types/Journal article
000286250 3367_ $$0PUB:(DE-HGF)16$$2PUB:(DE-HGF)$$aJournal Article$$bjournal$$mjournal$$s1702898982_30633
000286250 3367_ $$2BibTeX$$aARTICLE
000286250 3367_ $$2ORCID$$aJOURNAL_ARTICLE
000286250 3367_ $$00$$2EndNote$$aJournal Article
000286250 520__ $$aArtificial intelligence (AI) has a multitude of applications in cancer research and oncology. However, the training of AI systems is impeded by the limited availability of large datasets due to data protection requirements and other regulatory obstacles. Federated and swarm learning represent possible solutions to this problem by collaboratively training AI models while avoiding data transfer. However, in these decentralized methods, weight updates are still transferred to the aggregation server for merging the models. This leaves the possibility for a breach of data privacy, for example by model inversion or membership inference attacks by untrusted servers. Somewhat-homomorphically-encrypted federated learning (SHEFL) is a solution to this problem because only encrypted weights are transferred, and model updates are performed in the encrypted space. Here, we demonstrate the first successful implementation of SHEFL in a range of clinically relevant tasks in cancer image analysis on multicentric datasets in radiology and histopathology. We show that SHEFL enables the training of AI models which outperform locally trained models and perform on par with models which are centrally trained. In the future, SHEFL can enable multiple institutions to co-train AI models without forsaking data governance and without ever transmitting any decryptable data to untrusted servers.
000286250 536__ $$0G:(DE-HGF)POF4-313$$a313 - Krebsrisikofaktoren und Prävention (POF4-313)$$cPOF4-313$$fPOF IV$$x0
000286250 588__ $$aDataset connected to CrossRef, PubMed, , Journals: inrepo02.dkfz.de
000286250 650_7 $$2Other$$aArtificial intelligence
000286250 650_7 $$2Other$$aFederated learning
000286250 650_7 $$2Other$$aHistopathology
000286250 650_7 $$2Other$$aHomomorphic encryption
000286250 650_7 $$2Other$$aPrivacy-preserving deep learning
000286250 650_7 $$2Other$$aRadiology
000286250 7001_ $$aTayebi Arasteh, Soroosh$$b1
000286250 7001_ $$aSaldanha, Oliver Lester$$b2
000286250 7001_ $$aMüller-Franzes, Gustav$$b3
000286250 7001_ $$aKhader, Firas$$b4
000286250 7001_ $$aQuirke, Philip$$b5
000286250 7001_ $$aWest, Nicholas P$$b6
000286250 7001_ $$aGray, Richard$$b7
000286250 7001_ $$aHutchins, Gordon G A$$b8
000286250 7001_ $$aJames, Jacqueline A$$b9
000286250 7001_ $$aLoughrey, Maurice B$$b10
000286250 7001_ $$aSalto-Tellez, Manuel$$b11
000286250 7001_ $$0P:(DE-He78)90d5535ff896e70eed81f4a4f6f22ae2$$aBrenner, Hermann$$b12$$udkfz
000286250 7001_ $$aBrobeil, Alexander$$b13
000286250 7001_ $$0P:(DE-He78)b9e439a1aa1244925f92d547c0919349$$aYuan, Tanwei$$b14$$udkfz
000286250 7001_ $$0P:(DE-He78)c259d6cc99edf5c7bc7ce22c7f87c253$$aChang-Claude, Jenny$$b15$$udkfz
000286250 7001_ $$0P:(DE-He78)6c5d058b7552d071a7fa4c5e943fff0f$$aHoffmeister, Michael$$b16$$udkfz
000286250 7001_ $$aFoersch, Sebastian$$b17
000286250 7001_ $$aHan, Tianyu$$b18
000286250 7001_ $$aKeil, Sebastian$$b19
000286250 7001_ $$aSchulze-Hagen, Maximilian$$b20
000286250 7001_ $$aIsfort, Peter$$b21
000286250 7001_ $$aBruners, Philipp$$b22
000286250 7001_ $$aKaissis, Georgios$$b23
000286250 7001_ $$aKuhl, Christiane$$b24
000286250 7001_ $$aNebelung, Sven$$b25
000286250 7001_ $$aKather, Jakob Nikolas$$b26
000286250 773__ $$0PERI:(DE-600)1497450-2$$a10.1016/j.media.2023.103059$$gVol. 92, p. 103059 -$$p103059$$tMedical image analysis$$v92$$x1361-8415$$y2023
000286250 909CO $$ooai:inrepo02.dkfz.de:286250$$pVDB
000286250 9101_ $$0I:(DE-588b)2036810-0$$6P:(DE-He78)90d5535ff896e70eed81f4a4f6f22ae2$$aDeutsches Krebsforschungszentrum$$b12$$kDKFZ
000286250 9101_ $$0I:(DE-588b)2036810-0$$6P:(DE-He78)b9e439a1aa1244925f92d547c0919349$$aDeutsches Krebsforschungszentrum$$b14$$kDKFZ
000286250 9101_ $$0I:(DE-588b)2036810-0$$6P:(DE-He78)c259d6cc99edf5c7bc7ce22c7f87c253$$aDeutsches Krebsforschungszentrum$$b15$$kDKFZ
000286250 9101_ $$0I:(DE-588b)2036810-0$$6P:(DE-He78)6c5d058b7552d071a7fa4c5e943fff0f$$aDeutsches Krebsforschungszentrum$$b16$$kDKFZ
000286250 9131_ $$0G:(DE-HGF)POF4-313$$1G:(DE-HGF)POF4-310$$2G:(DE-HGF)POF4-300$$3G:(DE-HGF)POF4$$4G:(DE-HGF)POF$$aDE-HGF$$bGesundheit$$lKrebsforschung$$vKrebsrisikofaktoren und Prävention$$x0
000286250 9141_ $$y2023
000286250 915__ $$0StatID:(DE-HGF)0100$$2StatID$$aJCR$$bMED IMAGE ANAL : 2022$$d2023-10-21
000286250 915__ $$0StatID:(DE-HGF)0200$$2StatID$$aDBCoverage$$bSCOPUS$$d2023-10-21
000286250 915__ $$0StatID:(DE-HGF)0300$$2StatID$$aDBCoverage$$bMedline$$d2023-10-21
000286250 915__ $$0StatID:(DE-HGF)0600$$2StatID$$aDBCoverage$$bEbsco Academic Search$$d2023-10-21
000286250 915__ $$0StatID:(DE-HGF)0030$$2StatID$$aPeer Review$$bASC$$d2023-10-21
000286250 915__ $$0StatID:(DE-HGF)0199$$2StatID$$aDBCoverage$$bClarivate Analytics Master Journal List$$d2023-10-21
000286250 915__ $$0StatID:(DE-HGF)0113$$2StatID$$aWoS$$bScience Citation Index Expanded$$d2023-10-21
000286250 915__ $$0StatID:(DE-HGF)0150$$2StatID$$aDBCoverage$$bWeb of Science Core Collection$$d2023-10-21
000286250 915__ $$0StatID:(DE-HGF)0160$$2StatID$$aDBCoverage$$bEssential Science Indicators$$d2023-10-21
000286250 915__ $$0StatID:(DE-HGF)1160$$2StatID$$aDBCoverage$$bCurrent Contents - Engineering, Computing and Technology$$d2023-10-21
000286250 915__ $$0StatID:(DE-HGF)9910$$2StatID$$aIF >= 10$$bMED IMAGE ANAL : 2022$$d2023-10-21
000286250 9201_ $$0I:(DE-He78)C070-20160331$$kC070$$lC070 Klinische Epidemiologie und Alternf.$$x0
000286250 9201_ $$0I:(DE-He78)C120-20160331$$kC120$$lPräventive Onkologie$$x1
000286250 9201_ $$0I:(DE-He78)HD01-20160331$$kHD01$$lDKTK HD zentral$$x2
000286250 9201_ $$0I:(DE-He78)C020-20160331$$kC020$$lC020 Epidemiologie von Krebs$$x3
000286250 980__ $$ajournal
000286250 980__ $$aVDB
000286250 980__ $$aI:(DE-He78)C070-20160331
000286250 980__ $$aI:(DE-He78)C120-20160331
000286250 980__ $$aI:(DE-He78)HD01-20160331
000286250 980__ $$aI:(DE-He78)C020-20160331
000286250 980__ $$aUNRESTRICTED