Stage-oe-small.jpg

Thema4858/en

Aus Aifbportal
Wechseln zu:Navigation, Suche


Umbrella Review on Federated Learning in Healthcare



Information on the Thesis

Type of Final Thesis: Master
Supervisor: Florian Leiser
Research Group: Critical Information Infrastructures

Archive Number: 4.858
Status of Thesis: In Progress
Date of start: 2022-07-01
Date of submission: 2023-01-01

Further Information

Background:

Despite the ever-growing importance of machine learning (ML) in recent years, there are few ML solutions in healthcare compared to other sectors. Among many other concerns, data protection plays a particularly important role in this sector. Federated Learning (FL) seems to offer a solution to this by sharing only the weights of ML models between institutions, but not the data itself. To keep research up to date, surveys are regularly produced to collect and summarize progress in FL in healthcare. However, even most of these surveys focus on only one application area.


Objective(s):

The aim of this work is to create a so-called umbrella survey. The findings of existing surveys are to be summarized in order to identify central problems, goals and next steps. In this thesis such a "survey of surveys" for Federated Learning solutions in healthcare will be conducted. Thus, the literature must first be examined for existing reviews in order to later pool them and generate our own findings.


Literature:

  • Smith, V., Devane, D., Begley, C.M. et al. Methodology in conducting a systematic review of systematic reviews of healthcare interventions. BMC Med Res Methodol 11, 15 (2011). https://doi.org/10.1186/1471-2288-11-15
  • Thomson, D., Russell, K., Becker, L., Klassen, T. and Hartling, L. (2010), The evolution of a new publication type: Steps and challenges of producing overviews of reviews. Res. Synth. Method, 1: 198-211. https://doi.org/10.1002/jrsm.30
  • Rieke, N., Hancox, J., Li, W. et al. The future of digital health with federated learning. npj Digit. Med. 3, 119 (2020). https://doi.org/10.1038/s41746-020-00323-1