This work is part of the ELLIIT project "Federated Learning for Heterogeneous and Inconsistent Data", specifically focusing on the development of novel robust local models and methods for fusion of multimodal data, with this work conducted mainly in Lund. More information about the ELLIIT project is available at:
https://elliit.se/project/federated-learning-for-heterogeneous-and-inconsistent-data/
We will work with medical imaging data, including MRI and CT, and develop and evaluate machine learning models for medical image analysis. The research will focus on designing robust models and methods for combining information from heterogeneous and multimodal data sources. The work will include collaboration with researchers in radiation physics in Lund, providing additional medical imaging applications and expertise. We will also investigate the use of foundation models for medical imaging, including their adaptation, fine-tuning, and, where appropriate, training on relevant datasets.