Application of Federated Learning to Medical Data at Large Scale

Federated learning is a technique which allows training machine learning models in a distributed way without transferring the data from its source. It has thus potential applications in analysis of medical data, where privacy and security issues are of great importance. Although there are examples of using federated approaches to analysis of medical data, there is still need for research in this area and for experiments in real distributed environments.

This thesis aims to explore the application of federated learning techniques to solve specific problems in medical data analysis, such as MRI translation and estimation of brain microstructural parameters. The project will use publicly available datasets with varying acquisition parameters and resolution. The research will be carried out using a specific library for federated learning called Flower and HPC resources from PL-Grid.

The main objective of the research is to analyze the different federated learning techniques and determine their potential and limitations when applied to medical imaging. Additionally, the research aims to identify the necessary steps to make these methods more accessible and practical for real-world applications.