The formula for launching a machine learning company in health care looks something like this: Build a model, test it on historical patient data in a computer lab, and then start selling it to hospitals nationwide.

Suchi Saria, director of the machine learning and health care lab at Johns Hopkins University, is taking a different approach. Her company, Bayesian Health, is coming out of stealth mode on Monday by publishing a prospective study on how one of its lead products — an early warning system for sepsis — impacted the care of current patients in real hospitals.