Federated Learning and Integration of Biomedical Data

Explore federated learning and integration techniques for biomedical data in this informative lecture presented by Dr. Jose-Luis Ambite from USC Information Sciences Institute. Discover how hospitals, clinics, and researchers can collaborate on machine learning projects while maintaining data privacy and complying with regulations. Learn about the FLINT architecture, which combines federated learning with data integration methods to address challenges in distributed biomedical data analysis. Gain insights into encrypted model training, gradient noise protection, and the importance of data harmonization and imputation in neuroimaging tasks. Understand the speaker's expertise in data integration, databases, semantic web, and biomedical data science as you delve into this cutting-edge approach to collaborative medical research.