The Multiscale Biomedical Systems group leverages AI, advanced simulations, and high-performance computing to accelerate innovation in biomedical science. The group’s mission is to develop predictive models that connect biology across scales — from atoms and molecules to tissues, organs, and entire populations. By bridging these layers, the group drives breakthroughs in cancer therapy, drug discovery, biomedical imaging, and global health challenges.
The group’s work focuses on three research directions:
AI for Health and Therapeutics
Advancing precision medicine through intelligent systems for oncology, radiation treatment planning, biomedical imaging, and clinical decision support — Current efforts include the development of digital twins, virtual trials, and adaptive treatment workflows.
Molecular Design, Simulation, and Translational Impact
Designing novel therapeutics, smart materials, and functional polymers by leveraging generative AI and bridging atomic- to organ-level biology via multiscale simulation and experimental data integration — These tools enable real-world impact in drug discovery, disease modeling, and next-generation therapy development.
Foundation Models for Science and Engineering
Developing transformer-based AI models for diverse applications, including imaging, climate modeling, materials science, and sustainable agriculture — The group leverages platforms to build scalable, generalizable AI systems that drive discovery across domains. One such model, trained on Frontier, is the Oak Ridge Base Foundation Model for Earth System Predictability, the world’s largest AI model for weather prediction.
