Supervised Learning for COVID Mortality Span Prediction

Abstract

Abstract: The covid-19 outbreak is causing concern among the world's population. Because there are no preventative precautions in place, current treatment approaches are limited to treating patients who tested positive for covid-19. In this case, determining the severity of the patient's disease is critical in lowering the covid-19-related death rate. The pathology findings are used by clinical professionals to scale the severity of the condition. Clinical specialists' diagnosis is strongly dependent on their level of competence. Unlike the other dimensions, sensitivity, or accuracy in disease-prone situations, is extremely important in clinical practice. This paper describes a supervised learning strategy for performing computer-assisted covid-19 mortality scope using the target patient's pathology records. The experimental investigation shows the utility of the proposed strategy in predicting death with the least amount of false alerting.

 Index Terms: COVID-19, Computed Tomography, Feature Optimization, World Health Organization, Machine Learning.