Clinical management of long Covid cases: application of artificial intelligence techniques

• Grant number: 24/17780-3
• Support Opportunities: Scholarships in Brazil - Scientific Initiation
• Start date: April 01, 2025
• End date: March 31, 2026
• Field of knowledge: Health Sciences - Collective Health - Public Health
• Principal Investigator: Sílvia Carla da Sílva André Uehara
• Grantee: Júlia Hellen Ferreira de Sousa
• Host Institution: Centro de Ciências Biológicas e da Saúde (CCBS). Universidade Federal de São Carlos (UFSCAR). São Carlos , SP, Brazil

Abstract

This study aims to use artificial intelligence techniques to identify managementclinic for cases of post-Covid conditions. This is an analytical cross-sectional study and will becarried out in SUS reference hospitals for Covid-19 assistance in São Carlos-SP, in addition to the EpidemiologicalSurveillance Services and the Department of Basic and Outpatient Care. The study population will be made up of people whohad laboratory-diagnosed Covid-19, from January 2021 to December 31, 2023. It is not common practice for studies thatapply AI techniques to use sample calculation, as they analyze the basis official data or other studies; however, in thisproject, due to the lack of literature on the subject and an official database, it became necessary to use a database thatperformed a sample calculation, resulting in a sample of 349 cases of post-Covid conditions. The collected data will berecorded in spreadsheets; and, later, they will be organized in a database. The analysis of the variables collected in the studywill initially be exploratory in nature; Subsequently, a comprehensive evaluation of several Machine Learning (ML) algorithmswill be carried out to identify assistance in cases of post-Covid conditions.Furthermore, ML algorithms will be applied to predict other clinical management variables and groups with greaterpredisposition. Among the selection criteria, the explainability of the generated models will be considered, giving preferencetoalgorithms that are capable of generating more interpretable models, such as decision trees.