07/07/2026
Ottawa, Ontario — Tuesday July 7, 2026
Identifying which children are most likely to return to the emergency department with asthma isn’t always easy—but a new AI tool may be able to help.
Led by Dhenuka Radhakrishnan and Khaled El Emam, researchers at the CHEO Research Institute developed and tested AIRE-KIDS, a first of its kind for asthma care in Canadian children AI tool designed to predict which children who visit the emergency department with asthma are at highest risk of a future asthma-related visit or hospital admission.
“For many children with asthma, an emergency department visit isn’t a one-time event. Our study shows the potential for AI to help identify who may need more proactive asthma care so we can connect families with specialized support sooner.” – Dr. Dhenuka Radhakrishnan
Comparing multiple AI approaches, the team found that a specialized machine learning model developed and validated using electronic health records of nearly 4,000 children who received asthma care at CHEO delivered the most accurate predictions and showed promise as a tool to support referral decisions in the emergency department.
Researchers found that previous asthma-related emergency visits, medical complexity, asthma severity, food allergies, age, and prior respiratory-related emergency visits were among the strongest predictors of future asthma emergencies identified by AIRE-KIDS.
By helping clinicians more accurately identify children at highest risk, AIRE-KIDS could support earlier referrals to asthma education and specialist care—potentially helping families better manage asthma before another serious attack occurs.
Future studies will explore how the tool could be integrated into emergency department workflows to support real-time clinical decision-making.