Prediction of length of hospital stay in head and neck surgery using artificial intelligence.

• Grant number: 24/08464-0
• Support Opportunities: Scholarships in Brazil - Scientific Initiation
• Start date: July 01, 2024
• End date: June 30, 2025
• Field of knowledge: Health Sciences - Medicine - Surgery
• Principal Investigator: Luiz Paulo Kowalski
• Grantee: Willian Nogueira Silva
• Host Institution: Hospital das Clínicas da Faculdade de Medicina da USP (HCFMUSP). Secretaria da Saúde (São Paulo - Estado). São Paulo , SP, Brazil

Abstract

The length of hospital stay is an important indicator in the context of health, both for patientsand for the health system. This length of stay is defined as the duration of hospitalizationfrom the moment of admission to hospital discharge. Thus, it is a fundamental indicator fordetermining different treatment policies and resource allocation, demonstrating the relevanceof this information for patients and for the health system. Therefore, this meansunderstanding the amount of time required for care, in addition to being able to improve thelogistics of occupying beds and the order of surgical procedures. In head and neck surgery,which has important social relevance due to the high prevalence of new cases in Brazil,advance information about the length of hospital stay can make the allocation of resourcesmore pragmatic and improve the logistical management of the surgical procedures that willbe performed, contemplating the prediction of this variable. The present study seeks to makeuse of already used algorithms that worth Artificial Intelligence to predict the length ofhospital stay in head and neck surgery patients through preoperative variables. Thus, havinga retrospective cohort with data from the medical records of patients from the Hospital dasClínicas of the University of São Paulo and the Cancer Institute of the State of São Paulowho underwent head and neck surgery, in which data from the patients' medical records willbe analyzed and processed by Artificial Intelligence. Thus, the study will result in thecreation of a model capable of predicting the length of hospital stay based on thepreoperative data of these patients.