A new project at Purdue University that combines many types of available data in a “digital twin” model of bladder cancer may prove powerful enough to predict patient outcomes.