The Information Fusion Lab focuses on machine learning for multimodal, heterogeneous data, particularly from the health domain. Specifically, the lab researches techniques that can combine images, time series and structured data, and can introduce domain-specific saliency into the models. The Information Fusion lab encompasses expertise on a wide range of ML methods, including deep learning, ensemble learning, object recognition, transfer of causal models and normalizing flows. The target applications include modeling trajectories for chronic conditions such as Alzheimer’s disease and osteoarthritis, medical imaging, clinical outcome prediction for critical care applications and assurance of fairness in health across multiple populations.