Automatic Forecast of Intensive Care Unit Admissions: The Experience During the COVID-19 Pandemic in Italy

• DOI: 10.1007/s10916-023-01982-9
• Publication Date: 2023-08-05T08:01:47Z
• Authors: Danila Azzolina; Corrado Lanera; Rosanna Comoretto; Andrea Francavilla; Paolo Rosi; Veronica Casotto; Paolo Navalesi; Dario Gregori
• Keywords: Pandemic; Exponential Smoothing

Abstract The experience of the COVID-19 pandemic showed importance timely monitoring admissions to ICU admissions. ability promptly forecast epidemic impact on occupancy beds in is a key issue for adequate management health care system. Despite this, most literature predictive models Italy has focused predicting number infections, leaving trends ordinary hospitalizations and occupancies background. This work aims present an ETS approach (Exponential Smoothing Time Series) time series forecasting tool based models. results model are presented regions affected by epidemic, such as Veneto, Lombardy, Emilia-Romagna, Piedmont. mean absolute percentage errors (MAPE) between observed predicted remain lower than 11% all considered geographical areas. In this epidemiological context, proposed could be suitable monitor, manner, disease system, not only during early stages but also vaccination campaign, quickly adapt possible preventive interventions.