Using machine learning to forecast peak health care service demand in real-time during the 2022–23 winter season: A pilot in England, UK DOI Creative Commons
Roger Morbey, Dan Todkill,

Phil Moura

и другие.

PLoS ONE, Год журнала: 2025, Номер 20(1), С. e0292829 - e0292829

Опубликована: Янв. 27, 2025

During winter months, there is increased pressure on health care systems in temperature climates due to seasonal increases respiratory illnesses. Providing real-time short-term forecasts of the demand for services helps managers plan their services. Winter 2022–23 we piloted a new forecasting pipeline, using existing surveillance indicators which are sensitive syncytial virus (RSV). Indicators including telehealth cough calls and emergency department (ED) bronchiolitis attendances, both children under 5 years. We utilised machine learning techniques train select models that would best forecast timing intensity peaks up 28 days ahead. Forecast uncertainty was modelled usings novel generalised additive model location, scale shape (gamlss) approach enabled prediction intervals vary according level activity. The atypical because healthcare exceptionally high, RSV circulating community concerns around invasive group A streptococcal (iGAS) infections. However, our proved be adaptive higher peak once increasing iGAS started. Thus, have demonstrated utility approach, adding systems.

Язык: Английский

Using machine learning to forecast peak health care service demand in real-time during the 2022–23 winter season: A pilot in England, UK DOI Creative Commons
Roger Morbey, Dan Todkill,

Phil Moura

и другие.

PLoS ONE, Год журнала: 2025, Номер 20(1), С. e0292829 - e0292829

Опубликована: Янв. 27, 2025

During winter months, there is increased pressure on health care systems in temperature climates due to seasonal increases respiratory illnesses. Providing real-time short-term forecasts of the demand for services helps managers plan their services. Winter 2022–23 we piloted a new forecasting pipeline, using existing surveillance indicators which are sensitive syncytial virus (RSV). Indicators including telehealth cough calls and emergency department (ED) bronchiolitis attendances, both children under 5 years. We utilised machine learning techniques train select models that would best forecast timing intensity peaks up 28 days ahead. Forecast uncertainty was modelled usings novel generalised additive model location, scale shape (gamlss) approach enabled prediction intervals vary according level activity. The atypical because healthcare exceptionally high, RSV circulating community concerns around invasive group A streptococcal (iGAS) infections. However, our proved be adaptive higher peak once increasing iGAS started. Thus, have demonstrated utility approach, adding systems.

Язык: Английский

Процитировано

0