Work

A semi-mechanistic modeling strategy for infectious diseases forecasting: Error correction and probabilistic prediction

Year: 2025

Type: article

Abstract: • Scientific questions Infectious disease forecasting faces increasingly complex challenges due to climate change and technological advancements, which heighten viral transmission risks and complicate disease spread dynamics. A key challenge lies in developing infectious disease modeling that balances mechanistic interpretability with data-driven flexibility to improve forecast accuracy and quantify uncertainty. • Evidence before this study Current infectious disease modeling uses two paradigms: mechanistic and data-driven models. Compartmental models, classical tools for modeling infectious disease dynamics, are good at theoretical interpretability but can’t precisely handle real-world uncertainties and complex transmission dynamics. Deep learning can capture complex patterns from heterogeneous data but offers limited epidemiological insight. However, the limited interpretability of current deep learning approaches in transmission mechanisms restricts their use for mechanistic analysis. • New findings Our semi-mechanistic infectious diseases forecasting framework significantly enhances long-term prediction capability while maintaining a certain degree of model interpretability. By extracting latent signals from extensive meteorological and population mobility datasets, the model supplements conventional surveillance data with critical transmission dynamics signatures while quantifying outbreak uncertainty. These advances improve prediction accuracy by over 50%. In addition, our study shows that population mobility and meteorological factors significantly affect SEIRD model prediction errors. Frequent travel complicates the transmission of infectious diseases and brings more uncertainty, making it difficult for traditional dynamic prediction models to accurately forecast epidemics. • Significance of the study The semi-mechanistic infectious diseases forecasting framework preserves mechanistic interpretability while utilizing deep learning to capture complex mobility-weather-disease interactions, enabling more accurate and actionable epidemic forecasts. Global climate change and technological advancements have intensified the threats of pandemics, while complex transmission dynamics challenge infectious disease forecasting. Traditional compartmental models struggle to fully capture both the dynamic transmission processes and their associated uncertainties. Here, we develop a novel hybrid methodology that integrates dynamic modeling with statistical approaches, establishing a semi-mechanistic model for error correction and probabilistic prediction. Our error analysis of the dynamic model reveals that frequent population mobility compromises the accuracy of dynamic predictions, and that meteorological conditions further modulate forecast performance by regulating human movement patterns. To capture these effects, we implement a Quantile Regression Long Short-Term Memory network to estimate prediction errors of the epidemic dynamic model based on mobility and environmental data. This hybrid approach corrects dynamic prediction errors and generates probabilistic forecasts. Validation using multi-state the United States (U.S.) corona virus disease 2019 (COVID-19) outbreak data shows that our framework reduces dynamic prediction errors by over 50%. Compared with pure deep learning approaches, the semi-mechanistic model significantly enhances long-term prediction performance and interpretability. By integrating mechanistic modeling with data-driven learning, the proposed approach improves the predictive accuracy and reliability of models in real-world outbreaks, thereby delivering more effective decision support for public health interventions.

Source: Biosafety and Health

DOI: 10.1016/j.bsheal.2025.12.002

Authors: Zihan Hao, Jiaxuan Hu, Shujuan Hu, Zhen Zhang, Donghuai Jia, Jianping Huang

Institution: Lanzhou University

Language: English

Topic: Mathematical and Theoretical Epidemiology and Ecology Models

Subfield: Public Health, Environmental and Occupational Health

Field: Medicine

Domain: Health Sciences

SDG: Good health and well-being

Open Access status: gold

Estimated APC paid: $2,500