Improving Outbreak Forecasts Through Model Augmentation

2025-06-19

AI Summary

Existing epidemiological forecasting models—both empirical and mechanistic—exhibit markedly reduced accuracy during outbreak peaks, especially at the peak itself. To address this, we propose *epimodulation*, a novel framework that enables lightweight, plug-and-play, dynamic integration of domain-specific epidemiological knowledge (e.g., trends in the basic reproduction number ℛ₀ and transmission thresholds) with mainstream time-series models—including ARIMA, Holt-Winters, spline-based methods, and ensemble forecasts from the COVID-19 Forecast Hub. Crucially, this hybrid paradigm imposes dynamical constraints *during* prediction without requiring architectural modifications to the base models. Evaluated on COVID-19 and influenza hospitalization data, epimodulation improves overall forecast accuracy by 9.1% and 19.5%, respectively, and achieves substantially larger gains—20.7% and 25.4%—during peak outbreak periods. These results demonstrate marked enhancement in the robustness and reliability of short-term forecasts during critical public health response windows.

Technology Category

• Search and Optimization: Metareasoning and Metaheuristics
• Machine Learning: Mixture of Experts (MoE)
• Reasoning under Uncertainty: Relational Probabilistic Models

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

Accurate forecasts of disease outbreaks are critical for effective public health responses, management of healthcare surge capacity, and communication of public risk. There are a growing number of powerful forecasting methods that fall into two broad categories -- empirical models that extrapolate from historical data, and mechanistic models based on fixed epidemiological assumptions. However, these methods often underperform precisely when reliable predictions are most urgently needed -- during periods of rapid epidemic escalation. Here, we introduce epimodulation, a hybrid approach that integrates fundamental epidemiological principles into existing predictive models to enhance forecasting accuracy, especially around epidemic peaks. When applied to simple empirical forecasting methods (ARIMA, Holt--Winters, and spline models), epimodulation improved overall prediction accuracy by an average of 9.1% (range: 8.2--12.5%) for COVID-19 hospital admissions and by 19.5% (range: 17.6--23.2%) for influenza hospital admissions; accuracy during epidemic peaks improved even further, by an average of 20.7% and 25.4%, respectively. Epimodulation also substantially enhanced the performance of complex forecasting methods, including the COVID-19 Forecast Hub ensemble model, demonstrating its broad utility in improving forecast reliability at critical moments in disease outbreaks.

Problem

• Enhancing disease outbreak forecasts during rapid epidemic escalation
• Integrating epidemiological principles into existing predictive models
• Improving accuracy of COVID-19 and influenza hospitalization predictions

Innovation

• Hybrid approach integrating epidemiological principles
• Enhances forecasting accuracy during epidemic peaks
• Improves both simple and complex forecasting methods