Triage Prediction of a Real Dataset of COVID-19 Patients in Alava

Abstract

The COVID-19 pandemic has increased the pressure on developing clinical decision-making systems based on predictive algorithms, potentially helping to reduce the unmanageable strain on healthcare systems. In an attempt to address this challenging health situation, we attempted to provide a contribution to this endeavour with an in-depth study of a real-life dataset of covid-19 patients from a local hospital. In this paper, we approach the problem as triage prediction problem, formulated as multi-class classification problem, with special care on the age normalization of physiological variables. We report experimental results obtained on a data sample covering COVID-19 patients assisted in a local hospital. To do this, we tried to emulate the triage decisions of the physicians recorded in a dataset containing the measurements of physiological variables and the triage decision. We obtained results that provide encouragement for a real-life application development of the data balancing and classification in the prediction of the triage that the medical doctors will assign the critical patients.