Journal Article

DKFZ-2022-00845

http://join2-wiki.gsi.de/foswiki/pub/Main/Artwork/join2_logo100x88.png Swarm learning for decentralized artificial intelligence in cancer histopathology.

Saldanha, O. L. ; Quirke, P. ; West, N. P. ; James, J. A. ; Loughrey, M. B. ; Grabsch, H. I. ; Salto-Tellez, M. ; Alwers, E.DKFZ* ; Cifci, D. ; Ghaffari Laleh, N. ; Seibel, T. ; Gray, R. ; Hutchins, G. G. A. ; Brenner, H.DKFZ* ; van Treeck, M. ; Yuan, T.DKFZ* ; Brinker, T. J.DKFZ* ; Chang-Claude, J.DKFZ* ; Khader, F. ; Schuppert, A. ; Luedde, T. ; Trautwein, C. ; Muti, H. S. ; Foersch, S. ; Hoffmeister, M.DKFZ* ; Truhn, D. ; Kather, J. N.

2022
Nature America Inc. New York, NY

This record in other databases:  

Please use a persistent id in citations: doi:10.1038/s41591-022-01768-5

Abstract: Artificial intelligence (AI) can predict the presence of molecular alterations directly from routine histopathology slides. However, training robust AI systems requires large datasets for which data collection faces practical, ethical and legal obstacles. These obstacles could be overcome with swarm learning (SL), in which partners jointly train AI models while avoiding data transfer and monopolistic data governance. Here, we demonstrate the successful use of SL in large, multicentric datasets of gigapixel histopathology images from over 5,000 patients. We show that AI models trained using SL can predict BRAF mutational status and microsatellite instability directly from hematoxylin and eosin (H&E)-stained pathology slides of colorectal cancer. We trained AI models on three patient cohorts from Northern Ireland, Germany and the United States, and validated the prediction performance in two independent datasets from the United Kingdom. Our data show that SL-trained AI models outperform most locally trained models, and perform on par with models that are trained on the merged datasets. In addition, we show that SL-based AI models are data efficient. In the future, SL can be used to train distributed AI models for any histopathology image analysis task, eliminating the need for data transfer.

Note: 2022 Jun;28(6):1232-1239

---

Contributing Institute(s):

1. C070 Klinische Epidemiologie und Alternf. (C070)
2. Präventive Onkologie (C120)
3. DKTK HD zentral (HD01)
4. NWG Digitale Biomarker in der Onkologie (C140)
5. C020 Epidemiologie von Krebs (C020)

Research Program(s):

1. 313 - Krebsrisikofaktoren und Prävention (POF4-313) (POF4-313)

Appears in the scientific report 2022

---

Database coverage:
; BIOSIS Previews ; Biological Abstracts ; Clarivate Analytics Master Journal List ; Current Contents - Clinical Medicine ; Current Contents - Life Sciences ; Ebsco Academic Search ; Essential Science Indicators ; IF >= 80 ; JCR ; Nationallizenz ; SCOPUS ; Science Citation Index Expanded ; Web of Science Core Collection

---