Improving Detection of Disease Re-emergence Using a Web-Based Tool (RED Alert): Design and Case Analysis Study

Area of Science:

• Public Health
• Epidemiology
• Data Science

Background:

• Infectious disease re-emergence identification lacks consistent quantitative criteria, potentially hindering effective mitigation efforts.
• Assessing local and global disease re-emergence requires comprehensive data on incidence and contributing factors worldwide.
• Existing practices may not adequately address the complexities of tracking and understanding disease resurgences.

Purpose of the Study:

• To develop a tool for public health officials to detect and understand infectious disease re-emergence.
• To answer key questions regarding local disease re-emergence, its contributing factors, and global re-emergence potential.
• To provide a data-driven approach for proactive infectious disease surveillance.

Main Methods:

• Collected and integrated diverse disease-related data (case counts, vaccination rates, transmission indicators) from global health organizations.
• Developed a web-based tool, RED Alert, combining machine learning and visual analytics for disease re-emergence detection.
• Focused on four key diseases: measles, cholera, dengue, and yellow fever, evaluating model performance through case studies.

Main Results:

• Supervised learning models achieved 82%-90% accuracy in identifying local re-emergence events, with a manageable false positive rate (18%-31%).
• Case study reviews confirmed RED Alert's capability for local re-emergence detection.
• The tool provided actionable insights into factors driving local re-emergence and global trends.

Conclusions:

• RED Alert is the first tool specifically designed to address the challenges of infectious disease re-emergence.
• The tool demonstrates the feasibility of using machine learning for early detection and understanding of disease resurgences.
• RED Alert offers a novel approach to support public health decision-making in managing infectious disease threats.