Generative diffusion models for spatiotemporal influenza forecasting
PLOS Computational Biology · 2026
DOI: 10.1371/journal.pcbi.1014846 ↗Kurz gefasst
Fachartikel in der Zeitschrift PLOS Computational Biology, 2026. Verfasst von Joseph Lemaitre und 1 weiteren Autor:innen. Beteiligte Einrichtung: Department of Epidemiology, Gillings School of Global Public Health, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina, United…. Hauptthemen laut MEDLINE: Influenza, Human, Models, Statistical. Die Arbeit nennt 26 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: NIAID NIH HHS, National Institute of Health, Center for Forecasting and Outbreak Analytics.
Abstract (Originalsprache)
Forecasting infectious disease incidence can provide important information to guide public health planning, yet it is difficult because epidemic dynamics are complex. Current mechanistic and statistical approaches often struggle to capture multimodal uncertainty or emergent trends. Influpaint adapts denoising diffusion probabilistic models to epidemic forecasting. By encoding influenza seasons as spatiotemporal images in which pixel intensity represents incidence, Influpaint learns a rich distribution of disease dynamics from a hybrid dataset of surveillance and simulated trajectories. Forecasting is formulated as a conditional generation (inpainting) task from partial observations. We show that Influpaint generates realistic, diverse epidemic trajectories and achieves forecast accuracy that is competitive with leading ensemble methods in retrospective evaluation. In real-time evaluation during the 2022–2025 U.S. CDC FluSight challenges, performance improved substantially across seasons, with highly accurate but somewhat overconfident projections in 2024–2025. The best performance was achieved with a training dataset containing 30% surveillance and 70% simulated trajectories. These results show that diffusion models can capture important spatiotemporal structure in influenza dynamics and provide a flexible framework for probabilistic infectious disease forecasting.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Autor:innen und Institutionen
- Department of Epidemiology, Gillings School of Global Public Health, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina, United States of America.
- Department of Epidemiology, Johns Hopkins Bloomberg School of Public Health, Baltimore, Maryland, United States of America.
- Carolina Population Center, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina, United States of America.
ORCID-Profile: Joseph Lemaitre
Förderung
- NIAID NIH HHS
- National Institute of Health
- Center for Forecasting and Outbreak Analytics
Zitieren
APA
Lemaitre, J., & Lessler, J. (2026). Generative diffusion models for spatiotemporal influenza forecasting. PLOS Computational Biology. https://doi.org/10.1371/journal.pcbi.1014846
Vancouver
Lemaitre J, Lessler J. Generative diffusion models for spatiotemporal influenza forecasting. PLOS Computational Biology. 2026. doi:10.1371/journal.pcbi.1014846
BibTeX
@article{Lemaitre2026,
author = {Lemaitre, Joseph and Lessler, Justin},
title = {Generative diffusion models for spatiotemporal influenza forecasting},
journal = {PLOS Computational Biology},
year = {2026},
doi = {10.1371/journal.pcbi.1014846}
}
EUVIMED