Using artificial intelligence to advance pediatric infection prevention and control

Recent literature describes the deployment of different artificial intelligence (AI) technologies to potentially support infection prevention and control (IP&C) in both the community and healthcare environment. However, most studies focus on adults. This review explores the data and potential for AI to enhance IP&C for pediatric populations as well as recognizing important limitations.

In community settings, AI can educate families about infections and risk, recognize potential clusters and outbreaks of infectious pathogens, and prescreen individually infected patients prior to entering a healthcare facility. For admitted patients, AI has been used to identify patients at risk for healthcare-associated infections (HAIs) such as central line associated blood stream infections, and may assist infection preventionists in abstracting chart data for HAI surveillance. Limitations include potential biases in training data and the lack of prospective studies validating the use of AI for IP&C purposes, especially in heterogeneous pediatric populations.

AI can be a valuable tool in recognizing and controlling infections in both the community and healthcare settings. However, more studies in pediatric populations are needed, including prospective studies that validate tools created and trained on retrospective cohorts.