Journal Article

DZNE-2021-00468

http://join2-wiki.gsi.de/foswiki/pub/Main/Artwork/join2_logo100x88.png Swarm Learning for decentralized and confidential clinical machine learning.

Warnat-Herresthal, S. (First author)DZNE* ; Schultze, H. ; Shastry, K. L. ; Manamohan, S. ; Mukherjee, S. ; Garg, V. ; Sarveswara, R. ; Händler, K.DZNE* ; Pickkers, P. ; Aziz, A.DZNE* ; Ktena, S. ; Tran, F. ; Bitzer, M. ; Ossowski, S. ; Casadei, N. ; Herr, C. ; Petersheim, D. ; Behrends, U. ; Kern, F. ; Fehlmann, T. ; Schommers, P. ; Lehmann, C. ; Augustin, M. ; Rybniker, J. ; Altmüller, J. ; Mishra, N. ; Bernardes, J. P. ; Krämer, B. ; Bonaguro, L.DZNE* ; Schulte-Schrepping, J.DZNE* ; De Domenico, E.DZNE* ; Siever, C. ; Kraut, M.DZNE* ; Desai, M. ; Monnet, B. ; Saridaki, M. ; Siegel, C. M. ; Drews, A.DZNE* ; Nuesch Germano, M.DZNE* ; Theis, H.DZNE* ; Heyckendorf, J. ; Schreiber, S. ; Kim-Hellmuth, S. ; Study, C. A. (Collaboration Author) ; Nattermann, J. ; Skowasch, D. ; Kurth, I. ; Keller, A. ; Bals, R. ; Nürnberg, P. ; Rieß, O. ; Rosenstiel, P. ; Netea, M. G. ; Theis, F. ; Mukherjee, S.DZNE* ; Backes, M. ; Aschenbrenner, A. C.DZNE* ; Ulas, T.DZNE* ; Initiative, D. C. O. (Collaboration Author) ; Breteler, M.DZNE* ; Giamarellos-Bourboulis, E. J. ; Kox, M. ; Becker, M.DZNE* ; Cheran, S. ; Woodacre, M. S. ; Goh, E. L. ; Schultze, J. L.DZNE* ; Balfanz, P. ; Eggermann, T. ; Boor, P. ; Hausmann, R. ; Kuhn, H. ; Isfort, S. ; Stingl, J. C. ; Schmalzing, G. ; Kuhl, C. K. ; Röhrig, R. ; Marx, G. ; Uhlig, S. ; Dahl, E. ; Müller-Wieland, D. ; Dreher, M. ; Marx, N. ; Angelov, A. ; Bartholomäus, A. ; Becker, A. ; Bezdan, D. ; Blumert, C. ; Bonifacio, E. ; Bork, P. ; Boyke, B. ; Blum, H. ; Clavel, T. ; Colome-Tatche, M. ; Cornberg, M. ; De La Rosa Velázquez, I. A. ; Diefenbach, A. ; Dilthey, A. ; Fischer, N. ; Förstner, K. ; Franzenburg, S. ; Frick, J.-S. ; Gabernet, G. ; Gagneur, J. ; Ganzenmueller, T. ; Gauder, M. ; Geißert, J. ; Goesmann, A. ; Göpel, S. ; Grundhoff, A. ; Grundmann, H. ; Hain, T. ; Hanses, F. ; Hehr, U. ; Heimbach, A. ; Hoeper, M. ; Horn, F. ; Hübschmann, D. ; Hummel, M. ; Iftner, T. ; Iftner, A. ; Illig, T. ; Janssen, S. ; Kalinowski, J. ; Kallies, R. ; Kehr, B. ; Keppler, O. T. ; Klein, C. ; Knop, M. ; Kohlbacher, O. ; Köhrer, K. ; Korbel, J. ; Kremsner, P. G. ; Kühnert, D. ; Landthaler, M. ; Li, Y. ; Ludwig, K. U. ; Makarewicz, O. ; Marz, M. ; McHardy, A. C. ; Mertes, C. ; Münchhoff, M. ; Nahnsen, S. ; Nöthen, M. M.Extern* ; Ntoumi, F. ; Overmann, J. ; Peter, S. ; Pfeffer, K. ; Pink, I. ; Poetsch, A. R. ; Protzer, U. ; Pühler, A. ; Rajewsky, N. ; Ralser, M. ; Reiche, K. ; Ripke, S. ; da Rocha, U. N. ; Saliba, A.-E. ; Sander, L. E. ; Sawitzki, B. ; Scheithauer, S. ; Schiffer, P. ; Schmid-Burgk, J. ; Schneider, W. ; Schulte, E.-C. ; Sczyrba, A. ; Sharaf, M. L. ; Singh, Y. ; Sonnabend, M. ; Stegle, O. ; Stoye, J. ; Vehreschild, J. ; Velavan, T. P. ; Vogel, J. ; Volland, S. ; von Kleist, M. ; Walker, A. ; Walter, J. ; Wieczorek, D. ; Winkler, S. ; Ziebuhr, J.

2021
Nature Publ. Group London [u.a.]

This record in other databases:    

Please use a persistent id in citations: doi:10.1038/s41586-021-03583-3

Abstract: Fast and reliable detection of patients with severe and heterogeneous illnesses is a major goal of precision medicine1,2. Patients with leukaemia can be identified using machine learning on the basis of their blood transcriptomes3. However, there is an increasing divide between what is technically possible and what is allowed, because of privacy legislation4,5. Here, to facilitate the integration of any medical data from any data owner worldwide without violating privacy laws, we introduce Swarm Learning-a decentralized machine-learning approach that unites edge computing, blockchain-based peer-to-peer networking and coordination while maintaining confidentiality without the need for a central coordinator, thereby going beyond federated learning. To illustrate the feasibility of using Swarm Learning to develop disease classifiers using distributed data, we chose four use cases of heterogeneous diseases (COVID-19, tuberculosis, leukaemia and lung pathologies). With more than 16,400 blood transcriptomes derived from 127 clinical studies with non-uniform distributions of cases and controls and substantial study biases, as well as more than 95,000 chest X-ray images, we show that Swarm Learning classifiers outperform those developed at individual sites. In addition, Swarm Learning completely fulfils local confidentiality regulations by design. We believe that this approach will notably accelerate the introduction of precision medicine.

Keyword(s): Blockchain (MeSH) ; COVID-19: diagnosis (MeSH) ; COVID-19: epidemiology (MeSH) ; Clinical Decision-Making: methods (MeSH) ; Confidentiality (MeSH) ; Datasets as Topic (MeSH) ; Disease Outbreaks (MeSH) ; Female (MeSH) ; Humans (MeSH) ; Leukemia: diagnosis (MeSH) ; Leukemia: pathology (MeSH) ; Leukocytes: pathology (MeSH) ; Lung Diseases: diagnosis (MeSH) ; Machine Learning: trends (MeSH) ; Male (MeSH) ; Precision Medicine: methods (MeSH) ; Software (MeSH) ; Tuberculosis: diagnosis (MeSH)

---

Contributing Institute(s):

1. Population & Clinical Neuroepidemiology (AG Aziz)
2. United epigenomic platform (AG Schultze)
3. Population Health Sciences (AG Breteler 1)
4. Platform for Single Cell Genomics and Epigenomics at DZNE & University of Bonn (R&D PRECISE)
5. Epigenetics and Systems Medicine in Neurodegenerative Diseases (AG Fischer 1)
6. Statistics and machine learning (AG Mukherjee)

Research Program(s):

1. 354 - Disease Prevention and Healthy Aging (POF4-354) (POF4-354)
2. 352 - Disease Mechanisms (POF4-352) (POF4-352)

Experiment(s):

1. Rhineland Study / Bonn

Appears in the scientific report 2021

---

Database coverage:
; ; ; BIOSIS Previews ; Biological Abstracts ; Chemical Reactions ; Clarivate Analytics Master Journal List ; Current Contents - Agriculture, Biology and Environmental Sciences ; Current Contents - Life Sciences ; Current Contents - Physical, Chemical and Earth Sciences ; Ebsco Academic Search ; Essential Science Indicators ; IF >= 60 ; Index Chemicus ; JCR ; SCOPUS ; Science Citation Index Expanded ; Web of Science Core Collection ; Zoological Record

---