Federated learning for collaborative healthcare research and aging biomarkers
S.J.F. Garst (TU Delft - Electrical Engineering, Mathematics and Computer Science)
M.J.T. Reinders – Promotor (TU Delft - Electrical Engineering, Mathematics and Computer Science)
P. Eline Slagboom – Promotor (Leiden University Medical Center, Max Planck Institute for Biology of Ageing)
More Info
expand_more
Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.
Abstract
Ageing biomarkers are machine learning models that quantify an individual’s ageing process using biological data such as MRI scans or blood-based measurements. They offer the potential to detect accelerated ageing early, but their development requires large and diverse datasets—resources rarely available within a single medical institution. Federated learning (FL), which enables joint model training across sites without sharing raw data, provides a powerful solution for such collaborative biomedical research. This dissertation advances both the methodology of federated learning and the development of biological ageing biomarkers.