GeoSpreadAI

Problem. Infectious diseases continue to pose a significant threat to public health, particularly in regions with dynamic population mobility, limited surveillance capacity, and heterogeneous socio-environmental conditions. Traditional epidemiological models often fail to capture real-world complexity, including behavioural factors, spatial heterogeneity, and rapid changes in transmission pathways. Existing forecasting systems provide delayed or coarse-resolution insights, limiting their utility for early outbreak detection and timely intervention planning. As a result, health authorities lack precise, actionable, and data-driven tools to anticipate epidemic spread, allocate resources efficiently, and mitigate health-system burden during emerging infectious threats.

Main objective is to develop an AI-enabled multi-agent geospatial modelling platform capable of accurately simulating and forecasting infectious disease transmission in real-world environments. The objective is to integrate epidemiological, behavioural, mobility and environmental data into a unified predictive ecosystem that supports early outbreak detection, dynamic risk mapping, scenario development, and evidence-based public-health decision-making.

Expected outcomes. The project will deliver an AI-based multi-agent geospatial platform enabling high-resolution epidemic forecasting, early detection of transmission hotspots, and dynamic risk assessment. Expected outcomes include improved preparedness, more targeted and timely public-health interventions, optimised resource allocation, and strengthened resilience of health systems to emerging infectious threats.

What We Have Now? We have already developed and validated a functional multi-agent geospatial model using real epidemiological data on tuberculosis transmission. The system successfully reproduces spatial spread patterns, identifies high-risk clusters, and demonstrates strong predictive accuracy, confirming its readiness for scaling to other infectious diseases and broader public-health applications.

Our Expertise in Relation to the Topic:

•Infectious disease modelling & epidemiology:

Extensive research on tuberculosis and comorbid conditions, including analysis of transmission dynamics, risk factors, and clinical outcomes.

•AI-driven health analytics:

Experience applying machine learning, clustering, and predictive modelling to improve surveillance and early detection of infectious diseases.

•Multi-agent & geospatial simulation:

Development and successful validation of a geospatial multi-agent model for real-world TB spread, integrating behavioural and spatial heterogeneity.

•Clinical and laboratory expertise:

Strong background in respiratory medicine and access to high-quality patient datasets and diagnostic indicators for model calibration and validation.

•Research coordination & training:

Leadership in doctoral and postdoctoral programs, scientific project management, and national/international academic collaboration.

•Public-health integration:

Close cooperation with healthcare institutions to ensure practical applicability, policy relevance, and impact on health-system preparedness.