Visual analytics of geo-social interaction patterns for epidemic control

Area of Science:

• Epidemiology
• Computational Social Science
• Public Health

Background:

• Airborne disease spread is influenced by human interaction and population mobility.
• Geo-social interaction patterns, derived from mobility data, offer potential for pandemic mitigation.
• Limited research exists on control strategies designed using geo-social interaction patterns.

Purpose of the Study:

• Propose a framework for effective disease control by identifying geo-social patterns.
• Design and evaluate control measures based on these identified patterns.
• Develop a visual analytic tool to support disease control strategy design.

Main Methods:

• Developed a framework for disease control starting with pattern identification.
• Created a visual analytic tool integrating reorderable matrices, agent-based models, and visualizations.
• Used real-world human interaction data from a French primary school for a case study.

Main Results:

• Locally targeted vaccination strategies show potential in limiting infections and preventing regional spread.
• Varying spatial-social scales impact the success of local vaccination strategies.
• A proper spatial-social scale can enhance control efficacy with limited resources.

Conclusions:

• Geo-social patterns aid in targeting critical individuals, locations, and clusters for disease control.
• The GS-EpiViz tool supports designing and testing advanced control scenarios for airborne diseases.
• Prioritizing limited resources like vaccines can be improved by considering spatial-social scales.