Swarm Learning Technology
Swarm Learning for precision medicine in infectious diseases and pandemic preparedness
The increasing availability of high-resolution multi-omics and single-cell data is opening up new opportunities for AI-supported precision medicine, although some open questions remain regarding data protection, clinical applicability and ethical aspects. The Swarm Learning technology combines machine learning methods with a special kind of information exchange between study sites, in which AI algorithms are trained locally and all research data remains stored decentralized at the respective study site. Only algorithms and model parameters are exchanged, enabling research institutions to learn from each other without sharing sensitive patient data.
The project “Swarm Learning for precision medicine in infectious diseases and pandemic preparedness” is funded by the VW Foundation and brings together an interdisciplinary team of experts from the fields of data and information sciences, immunology, virology, infectiology, and ethics, under the lead of Prof. Joachim Schultze (DZNE). The project aims to establish a platform that uses the Swarm Learning technology to efficiently evaluate medical and immunological data, and centrally integrates them into new findings. To this end, a multicenter clinical study is being conducted to investigate the immune response after infection with SARS-CoV-2, influenza virus, and RSV.
Contact
Marie Weskamm
Links
DZNE, Swarm Learning Technologie
DZNE, press release on VW-sponsored swarm learning project
Publication
Warnat-Herresthal S, Schultze H, Shastry KL, et al. Swarm Learning for decentralized and confidential clinical machine learning. Nature. 2021 Jun;594(7862):265-270. doi: 10.1038/s41586-021-03583-3. Epub 2021 May 26.