A Hybrid ABM-PDE Framework for Real-World Infectious Disease Simulations

2025-03-11 · Source: cs.MA updates on arXiv.org · Field: Science & Research — Mathematics & Computational Sciences, Health & Medical Research, Research Methodology & Innovation · Depth: Expert, extended

Summary

A novel hybrid modeling framework couples an Agent-Based Model (ABM) with a Partial Differential Equation (PDE) model to simulate infectious disease spread in heterogeneous regions, specifically the Berlin-Brandenburg area. The framework models urban Berlin using a PDE and rural Brandenburg with an ABM, leveraging real-world mobile phone and infection data. This approach significantly reduces overall simulation runtime and achieves smaller errors compared to a full-ABM, requiring fewer runs for stable results (e.g., 29 runs for the hybrid model versus 92 for full-ABM at a 1% error threshold with a 25% population sample). The hybrid model enables efficient simulations for 100% of the population on standard hardware, demonstrating its practicality for large-scale epidemiological assessments.

Key takeaway

For computational epidemiologists or public health analysts modeling disease spread in diverse geographic areas, this hybrid ABM-PDE framework offers a compelling solution. You can achieve faster, more accurate simulations for large populations (e.g., 100%) on standard hardware, enabling quicker parameter fitting and more accessible real-time assessments than traditional full-ABM approaches. Consider its application for rapid policy evaluation and resource allocation.

Key insights

Hybrid ABM-PDE models enhance infectious disease simulation efficiency and accuracy in spatially diverse regions.

Principles

• Hybrid models reduce computational complexity.
• Region-specific modeling improves accuracy.
• Dynamic coupling ensures cross-domain consistency.

Method

The framework couples an ABM (for rural areas) with a PDE model (for urban areas), dynamically exchanging agents as density contributions at boundaries, and uses real-world mobility data for agent trajectories and landscape potential.

In practice

• Apply hybrid ABM-PDE for urban/rural regions.
• Incorporate real-world mobility data.
• Fit parameters for distinct time intervals.

Topics

• Hybrid Modeling
• Agent-Based Models
• Partial Differential Equations
• Epidemiological Modeling
• Spatial Dynamics
• Computational Efficiency

Code references

• matsim-org/matsim-episim-libs
• jlblancoc/nanoflann

Best for: Research Scientist, AI Scientist, Data Scientist

Related on AIssential

• DEpiABS: Differentiable Epidemic Agent-Based Simulator — DEpiABS is a differentiable agent-based model that balances mechanistic detail, computational efficiency, and interpretability for epidemic simulation.
• Integrating Mechanistic and Data-Driven Models for Neurological Disorders through Differentiable Programming — Hybrid modeling combines mechanistic and data-driven approaches to improve neurological disorder diagnostics and treatment.
• Constrained Hybrid Metaheuristic: A Universal Framework for Continuous Optimisation — cHM is a universal, adaptive hybrid metaheuristic that dynamically combines diverse optimizers for robust continuous optimization.
• Inference of latent epidemic regimes and generative simulations reveal how inequality and mobility shape COVID-19 transmission — Socioeconomic inequality and mobility jointly shape urban epidemic transmission dynamics and regime transitions.
• An Infectious Disease Spread Simulation Based on Large Language Model Decision Making — LLMs can simulate nuanced human health behaviors in spatially-grounded agent-based models, revealing demographic disparities.
• Hybrid Probabilistic Forecasting of Under-Five Malaria Admissions in Ghana: A Gaussian Process Regression with Holt-Winters Smoothing — Combining GPR with Holt-Winters provides robust probabilistic malaria forecasting, capturing non-linearity and seasonality.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by cs.MA updates on arXiv.org.