Chilean Avian flu and its marine impacts: an online Statistical Process Control task

2025-05-13

AI Summary

High-pathogenicity avian influenza virus (HPAIV) H5N1 has caused mass mortality events among marine wildlife—including seals and penguins—along the Pacific coast of South America, particularly in Chile, raising urgent concerns for wildlife conservation and zoonotic spillover risk. Method: We developed the first Bayesian hierarchical spatiotemporal model integrating spatial covariates with region-specific and shared spatial effects. The framework combines spatiotemporal statistical modeling, spatial autocorrelation estimation, and real-time statistical process control (SPC) to enable online monitoring, joint inference of transmission pathways and mortality hotspots, and dynamic tracking of outbreak evolution. Contribution/Results: Our novel incorporation of heterogeneous regional spatial effects into a Bayesian spatiotemporal framework significantly improves cross-regional transmission forecasting accuracy. The model precisely identifies high-risk hotspots in northern Chile, delivering actionable, data-driven early warnings and decision support for public health response and marine ecosystem protection.

Technology Category

• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding
• Data Mining & Knowledge Management: Mining of Spatial, Temporal or Spatio-Temporal Data

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• User Modeling, Personalization and Recommendation: User privacy protection in personalized systems
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs

Abstract

The rapid spread of the HPAI H5N1 virus, responsible for the Avian Flu, is causing a great catastrophe on the South American Pacific coast (especially in the south of Peru and north of Chile). Although very little attention has been delivered to this pandemic, it presents a tremendous lethal rate, though the number of infected humans is relatively low. Towards monitoring and statistical control, this work shows the Chilean national statistics from the year 2023, and presents the developed online tool for supporting the government's decision-making. Additionally, a Bayesian hierarchical spatiotemporal model was used to model the joint analysis of the weekly registered animal including spatial covariates as well as specific and shared spatial effects that take into account the potential autocorrelation between the HPAI H5N1 virus per Region. Our findings allow us to identify the hot-spot areas with high amounts of dead bodies (mostly pinnipeds and penguins) and their evolution over time.

Problem

• Monitoring HPAI H5N1 virus spread in Chilean marine ecosystems
• Developing online tool for government decision-making support
• Identifying hot-spot areas with high animal mortality rates

Innovation

• Online tool for government decision-making support
• Bayesian hierarchical spatiotemporal modeling approach
• Identification of hot-spot areas with spatial effects