Applications of prediction models in medicine
Much of our work is grounded in real applications, particularly in medicine, where good prediction can make a meaningful difference. We develop and evaluate models that use patient data to estimate risk, support clinical decisions, and ultimately improve health outcomes. In this space, it’s not enough for a model to perform well on paper—it has to be trustworthy, interpretable, and useful in practice. We work closely with clinical collaborators to ensure that our models address real needs and are evaluated in realistic settings. Think of this as taking our methodological “SOUP” from the kitchen to the clinic—making sure it’s not only well-crafted, but also nourishing and ready to serve where it’s needed most.