Forecasting seasonal influenza epidemics with physics-informed neural networks
5 affiliations
- 1Department of Information Engineering and Computer Science, University of Trento, Trento, Italy; Mobile and Social Computing (MobS) Lab, Fondazione Bruno Kessler (FBK), Trento, Italy. Electronic address: [email protected]
- 2Department of Environmental Sciences, Informatics and Statistics, University of Venice, Venice, Italy
- 3Centre de Physique Théorique, CNRS, Marseille, France
- 4Department of Sociology and Social Research, University of Trento, Trento, Italy
- 5Mobile and Social Computing (MobS) Lab, Fondazione Bruno Kessler (FBK), Trento, Italy
Abstract
Accurate epidemic forecasting is critical for informing public health decisions and timely interventions. While physics-informed neural networks have shown promise in various scientific domains, their potential application to real-time epidemic forecasting remains underexplored. Here, we present SIR-INN, a hybrid forecasting framework that integrates the mechanistic structure of the classical Susceptible-Infectious-Recovered (SIR) model into a neural network architecture. Trained once on synthetic epidemic scenarios, the model is able to generalize across epidemic conditions without retraining. From limited and noisy observations, SIR-INN infers key transmission parameters via Markov chain Monte Carlo, generating probabilistic short- and long-term forecasts. We validate SIR-INN using national influenza data from the Italian National Institute of Health in the 2023-2024 and 2024-2025 seasons. The model performs competitively with current state-of-the-art approaches, particularly in terms of weighted interval score. It shows accurate predictive performance in nearly all phases of the outbreak, with improved accuracy observed for the 2024-2025 influenza season. Credible uncertainty intervals are consistently maintained, while coverage metrics highlight room for improvement in uncertainty calibration. SIR-INN offers a computationally efficient, transparent, and generalizable solution for epidemic forecasting, appropriately leveraging the framework's hybrid design. Its ability to provide real-time predictions of epidemic dynamics, together with uncertainty quantification, makes it a promising tool for real-world epidemic forecasting.
Related
- Forecasting of infectious disease epidemics in sexual networks using attenuated renewal processes
- Sparse-Observation Atmospheric Thermal Forecasting with Physics-Informed Neural Networks for Climate-Aware Digital Twins
- Station-specific thunderstorm forecasting using climate predictors with seasonal decomposition analysis in Bangladesh’s most vulnerable regions
- Forecasting Weekly Influenza Detections in South Africa Using SARIMA and LSTM Models
- Generative diffusion models for spatiotemporal influenza forecasting
BibTeX
@article{rama2026forecasting,
title = {{Forecasting seasonal influenza epidemics with physics-informed neural networks}},
author = {Martina Rama and Gabriele Santin and Giulia Cencetti and Michele Tizzoni and Bruno Lepri},
journal = {Epidemics},
year = {2026},
month = may,
doi = {10.1016/j.epidem.2026.100919},
url = {https://doi.org/10.1016/j.epidem.2026.100919},
note = {CC BY-NC-ND},
}
RIS
TY - JOUR
TI - Forecasting seasonal influenza epidemics with physics-informed neural networks
AU - Martina Rama
AU - Gabriele Santin
AU - Giulia Cencetti
AU - Michele Tizzoni
AU - Bruno Lepri
PY - 2026
DA - 2026/05/29
JO - Epidemics
DO - 10.1016/j.epidem.2026.100919
UR - https://doi.org/10.1016/j.epidem.2026.100919
AB - Accurate epidemic forecasting is critical for informing public health decisions and timely interventions. While physics-informed neural networks have shown promise in various scientific domains, their potential application to real-time epidemic forecasting remains underexplored. Here, we present SIR-INN, a hybrid forecasting framework that integrates the mechanistic structure of the classical Susceptible-Infectious-Recovered (SIR) model into a neural network architecture. Trained once on synthetic epidemic scenarios, the model is able to generalize across epidemic conditions without retraining. From limited and noisy observations, SIR-INN infers key transmission parameters via Markov chain Monte Carlo, generating probabilistic short- and long-term forecasts. We validate SIR-INN using national influenza data from the Italian National Institute of Health in the 2023-2024 and 2024-2025 seasons. The model performs competitively with current state-of-the-art approaches, particularly in terms of weighted interval score. It shows accurate predictive performance in nearly all phases of the outbreak, with improved accuracy observed for the 2024-2025 influenza season. Credible uncertainty intervals are consistently maintained, while coverage metrics highlight room for improvement in uncertainty calibration. SIR-INN offers a computationally efficient, transparent, and generalizable solution for epidemic forecasting, appropriately leveraging the framework's hybrid design. Its ability to provide real-time predictions of epidemic dynamics, together with uncertainty quantification, makes it a promising tool for real-world epidemic forecasting.
SN - 1755-4365
SN - 1878-0067
ER -