Cluster Detection Mechanisms for Syndromic Surveillance Systems: Systematic Review and Framework Development

Area of Science:

• Public Health
• Computer Science
• Data Science

Background:

• Timely detection of disease outbreaks is crucial for global health security.
• Advancements in technology facilitate access to health data for syndromic surveillance.
• Clustering algorithms are central to disease surveillance for identifying outbreaks based on data similarities.

Purpose of the Study:

• To systematically review implemented disease surveillance clustering algorithms (temporal, spatial, spatiotemporal).
• To evaluate the usage and performance of these clustering mechanisms.
• To develop an efficient cluster detection mechanism framework.

Main Methods:

• Conducted a systematic literature review across multiple databases (Google Scholar, PubMed, Scopus, etc.).
• Included studies published in peer-reviewed journals focusing on practically implemented syndromic surveillance systems and infectious diseases.
• Developed a framework for cluster detection mechanisms based on the review findings.

Main Results:

• Identified and reviewed 27 relevant articles after an extensive search and filtering process.
• Confirmed that various clustering and aberration detection algorithms have been empirically implemented and tested with real data.
• Developed a comprehensive framework encompassing data processing, clustering, visualization, and alerting.

Conclusions:

• The review identified numerous practically implemented and tested algorithms for disease surveillance.
• These findings can guide the development of effective cluster detection mechanisms for syndromic surveillance.
• The proposed framework supports diverse spatiotemporal surveillance needs for improved outbreak detection.