Study of Ensemble Learners to identify the COVID-19 positive patients

ABSTRACT

SARS-CoV-2 coronavirus has already attracted substantial attention of the scientific community. Medical Science had never faced a tougher challenge than this pandemic. The rapid spread of the virus has caused a monumental increase in hospital admissions and deaths resulting in availability of data for analysis. Moreover, the disease has now become asymptomatic in most cases yet could be fatal for co-morbid patients. Patients on arrival to hospitals, whatever the case may be, are generally advised to opt for economically reasonable routine blood tests and certain aspects of this blood testing can assist us in determining if a patient is infected with coronavirus or not, at a very early stage. We can utilize ensemble classifiers (i.e., conglomerate of advanced and improved ensemble of learning algorithms) to distinguish between infected and non-infected individuals and rule out the scope of further spreading. In this paper, we have done a comparative study of the diverse ensemble learning techniques that are implemented over different patient's blood test reports and can presage if a patient is infected with coronavirus. © 2022 IEEE.