Federated Learning Across NCI Cancer Centers with FLAIMME
Are you attending the 2026 NVIDIA FLARETM Day? If so, be sure to check out the presentation from NCI CBIIT’s Dr. Umit Topaloglu at 12:10 p.m. Pacific Time (PT). He’ll discuss how NCI’s Federated Learning-based AI in Multimodal Exploration (FLAIMME) consortium is helping NCI-designated cancer centers share artificial intelligence (AI) models while protecting data privacy.
Specifically, he’ll discuss:
- the operational experience of FLAIMME.
- lessons about deployment timelines, failure modes, data harmonization, and governance of a multi-site NVIDIA FLARE infrastructure that spans six oncology use cases.
- how four of the NCI-designated cancer centers are seeking to improve clinical trial matching across their sites by federating MedGemma (a Gemma 3-based medical AI model).
So, whether you’re a data scientist interested in developing AI in oncology, or a cancer researcher seeking to learn more about these collaborative networks, attend this presentation to better understand the power of federated learning. And if you’re curious to learn more about NCI’s federated learning activities, visit our project page.
About the Speaker
Umit Topaloglu, Ph.D.
Dr. Topaloglu is the branch chief of NCI CBIIT’s Clinical & Translational Research Informatics Branch. In this position, he leads NCI’s clinical research informatics strategy across precision medicine, clinical trials reporting, and enterprise modernization. He is also the founder and leader of the FLAIMME consortium.