Predicting Inpatient Admissions in Brazilian Hospitals

• Author(s): Consoli, Bernardo ; Vieira, Renata ; Bordini, Rafael ; Manssour, Isabel
• Date: 2024
• Persistent ID: http://hdl.handle.net/10174/37132
• Origin: Repositório Científico da Universidade de Évora
• Document Type: Lecture
• Language: English

Description

Patient length-of-stay prediction is a topic of interest for hospital administrators, as it can aid in planning and the allocation of critical resources. Ideal resource allocation can result in better care and reduced costs. Artificial Intelligence solutions have been tested for this purpose using several datasets for both foreign and Brazilian hospitals, but focusing on long-term inpatient care or Intensive Care Unit patient flow. We propose using similar solutions to predict inpatient flow from common patient entry points, such as emergency care or walk-in appointments, in an effort to better understand whether a patient will require outpatient care or inpatient admission as early as possible. Our solution was able to predict inpatient flow with as much as 88% accuracy.

We gratefully acknowledge partial financial support by CNPq Scholarship - Brazil (projects 303208/2023-6 and 25/2020), FAPERGS 22/2551-0000390-7 (RITE CIARS Inteligência Artificial Aplicada à Saúde), Capes, and the FCT under projects CEECIND/01997/2017 and UIDB/00057/2020 (Portugal).