Artificial intelligence in infection surveillance: Data integration, applications and future directions

Area of Science:

• Public Health
• Infectious Disease Epidemiology
• Computational Epidemiology

Background:

• Traditional infection surveillance methods suffer from slow response times and limited analytical capabilities.
• Integrating diverse data sources is crucial for timely and accurate disease monitoring.
• Artificial Intelligence (AI) offers advanced analytical power to overcome these limitations.

Purpose of the Study:

• To review the transformative potential of AI in infection surveillance.
• To analyze AI's capacity to process heterogeneous datasets for improved disease detection and prediction.
• To synthesize specific AI applications in public health surveillance.

Main Methods:

• Narrative review of peer-reviewed articles and reports.
• Analysis of AI's machine learning capabilities for processing diverse data (EHRs, social media, spatiotemporal, wearables).
• Synthesis of AI applications in outbreak prediction, real-time monitoring, and transmission tracking.

Main Results:

• AI-driven models improve outbreak prediction and disease incidence estimation.
• AI enhances risk assessment by identifying susceptible populations and geographic hotspots.
• Specific applications like social media for influenza forecasting and wearables for real-time monitoring show promise.

Conclusions:

• AI significantly enhances infection surveillance, enabling earlier detection and improved public health strategies.
• Challenges include data privacy, model validation, and the need for diverse testing.
• Future work should focus on AI model adaptability, robust validation, and data integration for effective interventions.