Machine Learning and Laboratory Values in the Diagnosis, Prognosis and Vaccination Strategy of COVID-19

ABSTRACT

In March 2020, the World Health Organization (WHO) declared a pandemic status for COVID-19 disease caused by SARS-CoV-2 infection. Early diagnosis undoubtedly plays a fundamental role in the management of emergencies for the treatment of infected patients, whose prognosis may benefit from early treatment and containment of contagions in asymptomatic or pauci-symptomatic subjects. To date, the gold-standard technique for diagnosing SARS-CoV-2 infection is the identification of viral genomic material (RNA) by molecular diagnosis. To improve the pandemic management, the need for enhanced diagnostic capacity for SARS-CoV-2 infections soon emerged, with rapid, accurate, and easily accessible methods. Machine learning (ML) models could help define the diagnosis and, in some cases, even the prognosis of COVID-19 patients. This chapter describes ML models based on laboratory tests combined with other biometric parameters;the applications aimed at optimizing diagnosis and prognosis were mainly described. Finally, the vaccination campaign against SARS-CoV-2. The fields most considered were the heterogeneity in patient selection, laboratory parameters used, the machine learning models and their validation and implementation. Furthermore, we briefly describe artificial intelligence's potentialities in planning different strategies for the vaccination campaigns against SARS-COV-2 through laboratory tests. © The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Switzerland AG 2022.