Identifying outbreaks in sewer networks: An adaptive sampling scheme under network's uncertainty

Area of Science:

• Epidemiology
• Network Science
• Mathematical Modeling

Background:

• Wastewater surveillance is crucial for monitoring infectious diseases like COVID-19.
• Uncertainty in network structure poses challenges for accurate outbreak localization.

Purpose of the Study:

• To develop mathematical and algorithmic tools for identifying outbreak locations in uncertain sewer networks.
• To enable adaptive, day-by-day sampling for efficient outbreak detection.

Main Methods:

• Proposed a framework for adaptive sampling in uncertain networks.
• Utilized a heuristic strategy for sequential node testing.
• Validated the model on real and synthetic network data.

Main Results:

• Achieved univocal detection of infected nodes despite network uncertainty.
• Demonstrated that network uncertainty only slightly increases detection time.
• Showcased practical applicability of the methodology.

Conclusions:

• The developed framework effectively localizes outbreaks in wastewater systems.
• Adaptive sampling provides a robust solution for uncertain network conditions.
• The methodology is practical for real-world public health surveillance.