Artificial-Intelligence-Supported In-Hospital Cardiac Arrest Prediction, Prevention, and Management

In-hospital cardiac arrest is one of the most challenging and overlooked health issues worldwide. Even with care teams dedicated to rapid response when an arrest occurs, only about 1 in 5 patients survive. Since a person’s chance of survival goes down with every minute they are not treated after the arrest, we need to find ways to act before this downward spiral. As many of these cardiac arrests are preventable, if we could figure out which ones we should be able to prevent and how, we could improve patient survival much more than by treating them only after arrests occur.

We will use artificial intelligence to create models that predict and inform which patients are at the most risk of cardiac arrest, whether the arrest might be preventable, and what actions have the best chance of preventing the arrest.

With these models, we’ll be able to identify the unique risks and needs of each patient and respond with the evidence-based actions most likely to prevent a cardiac arrest. In addition, the models will help patients and their healthcare providers have better conversations about the goals of their care, and the kinds of choices they want to make.

Faculty Supervisor:

Rafid Mahmood

Student:

Partner:

University of Ottawa Heart Institute

Discipline:

Business

Sector:

Health and Related Sciences & Technology

University:

University of Ottawa

Program: