Artificial Intelligence Revolutionizes Healthcare-Associated Infection Surveillance

The National Institute for Public Health and the Environment in Bilthoven, Netherlands has conducted groundbreaking research on the application of artificial intelligence (AI) in healthcare-associated infection surveillance. The study found that AI-based methods have been applied less frequently in automated surveillance and more frequently for early prediction, particularly for sepsis. Despite heterogeneity in settings, populations, definitions, and model designs, AI-based models have shown promising results with moderate to very good performance and predicted sepsis within 0-40 hours before onset.

Key Takeaways:

• AI-based algorithms have been developed to identify patients with various healthcare-associated infections, including pneumonia, bloodstream, surgical site, catheter-associated urinary tract, and Clostridioides difficile infections, with a sensitivity of 54.2%-100% and specificity of 63.5%-100%.
• The use of large language models is expected to support and improve different aspects within the surveillance process, including more precise identification of patients with healthcare-associated infections.
• AI-based methods have been applied less frequently in automated surveillance, but have shown promising results with moderate to very good performance and predicted sepsis within 0-40 hours before onset.
• The implementation of AI-supported automated surveillance systems for healthcare-associated infections in daily practice remains scarce, and successful development and implementation demand meeting requirements related to technical capabilities, governance, practical, and regulatory considerations, and quality monitoring.
• The research emphasized the potential of AI in transforming healthcare-associated infection surveillance and prediction, offering more objective and timely infection rates and predictions.
• The study highlighted the need for exploring AI-based prediction models detecting patients at risk of developing different healthcare-associated infections further.

Statistics:

• Sensitivity of AI-based algorithms for identifying patients with various healthcare-associated infections: 54.2%-100%.
• Specificity of AI-based algorithms for identifying patients with various healthcare-associated infections: 63.5%-100%.
• Performance of AI-based models: moderate to very good (accuracy 61%-99%).
• Predicted sepsis within 0-40 hours before onset: 61%-99%.

Sources:

• "The future of healthcare-associated infection surveillance: Automated surveillance and using the potential of artificial intelligence." Journal of Internal Medicine, 2025.
• National Institute for Public Health and the Environment, Bilthoven, Netherlands.
• National Institute for Public Health and the Environment Reports: [1] NewsRx. National Institute for Public Health and the Environment Reports Findings in Artificial Intelligence (The future of healthcare-associated infection surveillance: Automated surveillance and using the potential of artificial intelligence). Journal of Engineering. June 16, 2025; p 1592.