SOLVAR: hybrid endemic-epidemic machine learning for early forecasting of high-incidence multidrug-resistant organism periods
Chiu, Edwin Kwan-Yeung, Wong, Shuk-Ching, Ip, Jonathan Daniel, So, Simon Yung-Chun, Yuen, Kwok-Yung, Chiu, Kelvin Hei-Yeung, Cheng, Vincent Chi-Chung npj Antimicrobials and Resistance
· Springer Nature
· 2026-11-18
· DOI: 10.1038/s44259-026-00300-w
Аннотация
Healthcare-associated multidrug-resistant organism (MDRO) outbreaks require proactive infection control, yet traditional surveillance remains reactive, and machine learning (ML) models ignore transmission dynamics. We developed Stochastic Outbreak Learning and Variance-Aware Regression (SOLVAR), a hybrid framework to classify high-incidence periods 1–3 weeks ahead. SOLVAR adaptively ensembles a direct ML classifier, an ML-informed endemic-epidemic (EE) expert, and a baseline EE expert, with weights learned from walk-forward validation performance. Using 44,386 incident episodes across five MDROs from a 1700-bed hospital (2021–2026), we benchmarked SOLVAR against ML-only, standalone EE, Poisson generalized linear model, autoregressive integrated moving average, and naive baselines using rolling origin walk-forward validation over 96 weeks. SOLVAR achieved the highest area under the receiver operating characteristic curve for carbapenem-resistant Enterobacterales (CRE) and multidrug-resistant Acinetobacter baumannii at every forecast horizon (up to 0.955 and 0.718), with balanced sensitivity and specificity for a 1-week CRE high-incidence period forecast (0.769 and 0.829; negative predictive value 0.906). For highly endemic methicillin-resistant Staphylococcus aureus , SOLVAR yielded a non-degenerate classifier where ML-only and standalone EE models failed. By combining ML and EE experts through performance-based adaptive weighting and explicitly modeling overdispersion, SOLVAR supports pathogen-specific classification of high-incidence periods 1–3 weeks ahead to inform infection control planning.
Chiu, Edwin Kwan-Yeung, Wong, Shuk-Ching, Ip, Jonathan Daniel et al. SOLVAR: hybrid endemic-epidemic machine learning for early forecasting of high-incidence multidrug-resistant organism periods. // npj Antimicrobials and Resistance. – 2026. – DOI: 10.1038/s44259-026-00300-w. – URL: https://doi.org/10.1038/s44259-026-00300-w
Chiu, Edwin Kwan-Yeung, Wong, Shuk-Ching, Ip, Jonathan Daniel, So, Simon Yung-Chun, Yuen, Kwok-Yung, Chiu, Kelvin Hei-Yeung, Cheng, & Vincent Chi-Chung (2026). SOLVAR: hybrid endemic-epidemic machine learning for early forecasting of high-incidence multidrug-resistant organism periods. npj Antimicrobials and Resistance. https://doi.org/10.1038/s44259-026-00300-w
Chiu, et al. “SOLVAR: hybrid endemic-epidemic machine learning for early forecasting of high-incidence multidrug-resistant organism periods.” npj Antimicrobials and Resistance, 2026. doi:10.1038/s44259-026-00300-w.
@article{Chiu2026,
title = {SOLVAR: hybrid endemic-epidemic machine learning for early forecasting of high-incidence multidrug-resistant organism periods},
author = {Chiu and Kwan-Yeung, Edwin and Wong and Shuk-Ching and Ip and Daniel, Jonathan and So and Yung-Chun, Simon and Yuen and Kwok-Yung and Chiu and Hei-Yeung, Kelvin and Cheng and Chi-Chung, Vincent},
journal = {npj Antimicrobials and Resistance},
year = {2026},
doi = {10.1038/s44259-026-00300-w},
url = {https://doi.org/10.1038/s44259-026-00300-w}
}
Background Diesel exhaust particles (DEP) are a major component of combustion-derived PM_2.5 and contain complex mixtures of organic and inorganic contaminants associated with elemental carbon (black carbon). Diesel particulate matter...
Background Triptolide (TP), a bioactive diterpenoid derived from Tripterygium wilfordii Hook F., exhibits potent anti-inflammatory and immunosuppressive activities but has a narrow therapeutic window and documented toxicities in...
Background Residents living near smelters and abandoned metal mines may experience chronic heavy metal exposure through multiple environmental pathways. Proteomic profiling provides a promising approach for identifying early biological...
Backgrounds Micro- and nanoplastics are associated with metabolic toxicity, but their effects on hepatic cholesterol homeostasis and potential mitigating interventions remain insufficiently characterized. Objectives To characterize...
Background Per- and polyfluoroalkyl substances (PFAS) are persistent environmental contaminants that accumulate in human tissues and may cause prolonged cellular and organ damage. Ferroptosis, an iron-dependent form of regulated cell...