Proper management of emergency rooms is needed to improve healthcare and patient satisfaction, guiding resource allocation. Predicting access and hospitalisation rates through Machine Learning appears feasible and promising, especially when coupled with air pollution and weather data. This work further investigates, in a more detailed way, a previously presented approach that applied predictive algorithms to data related to Brescia’s clinical and environmental data from 2018 to 2022 to predict daily accesses or daily hospitalisations for cardiovascular or respiratory disorders. Starting from the previous work, that analysis was improved and widened to a greater geographical area. The applied algorithms’ performances satisfactorily adhere to the actual data, especially when using the Support Vector Machine and Random Forest’s models as regressors on daily accesses and respiratory disease-caused hospitalisations. Even if the specific value is not always correctly predicted, generally, the overall trend seems to be rightly forecasted, and performance metrics are rather satisfying. Although additional work could still be encouraged to improve the models’ performances, results are rewarding and represent a new point of view on a complex and relevant matter. The real-life application of this One Health approach is now possible and could quite easily be adapted to other areas, too, with the final objective of improving the quality of healthcare and people’s quality of life.

Prediction of Emergency Department Visits Applying an One Health Approach: Further Investigations

Francesca Marinaro
Conceptualization
;
Andrea Buccoliero;
2024-01-01

Abstract

Proper management of emergency rooms is needed to improve healthcare and patient satisfaction, guiding resource allocation. Predicting access and hospitalisation rates through Machine Learning appears feasible and promising, especially when coupled with air pollution and weather data. This work further investigates, in a more detailed way, a previously presented approach that applied predictive algorithms to data related to Brescia’s clinical and environmental data from 2018 to 2022 to predict daily accesses or daily hospitalisations for cardiovascular or respiratory disorders. Starting from the previous work, that analysis was improved and widened to a greater geographical area. The applied algorithms’ performances satisfactorily adhere to the actual data, especially when using the Support Vector Machine and Random Forest’s models as regressors on daily accesses and respiratory disease-caused hospitalisations. Even if the specific value is not always correctly predicted, generally, the overall trend seems to be rightly forecasted, and performance metrics are rather satisfying. Although additional work could still be encouraged to improve the models’ performances, results are rewarding and represent a new point of view on a complex and relevant matter. The real-life application of this One Health approach is now possible and could quite easily be adapted to other areas, too, with the final objective of improving the quality of healthcare and people’s quality of life.
2024
Forecasting; ER accesses; Hospitalisations; Pollution; Weather; One Health; Environmental exposure
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11562/1153898
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
social impact