Use of machine learning for triage and transfer of ICU patients in the Covid-19 pandemic period: Scope Review

ObjectiveTo map, summarize and analyze the available studies on the use of artificial intelligence, for both triage and transfer of patients in intensive care units in situations of bed shortage crisis so that health teams and organizations make decisions based on updated technological tools of triage and transfer. MethodsScope review made in the databases Pubmed, Embase, Web of Science, CINAHL, Cochrane, LILACS, Scielo, IEEE, ACM and the novel Rayyan Covid database were searched. Supplementary studies were searched in the references of the identified primary studies. The time restriction is from 2020, and there was no language restriction. All articles aiming at the use of machine learning within the field of artificial intelligence in healthcare were included, as well as studies using data analysis for triage and reallocation of elective patients to ICU vacancies within the specific context of crises, pandemics, and Covid-19 outbreak. Studies involving readmission of patients were excluded. ResultsThe results excluded specific triage such as oncological patients, emergency room, telemedicine and non structured data. ConclusionMachine learning can help ICU triage, bed management and patient transfer with the use of artificial intelligence in situations of crisis and outbreaks. DescriptorsArtificial Intelligence. Machine learning. Intensive Care Units. Triage. Patient Transfer. COVID-19.