Using Bayesian Truncated Regression Model To Predict Thrombosis For COVID-19 Patients

Abstract

As it is clear that corona-virus (covid-19) has a direct danger to humanity in the world, from the beginning of appearing this virus till seven months later that medical science was unable to understand the behavior of this virus, in our study we focused on the Thrombosis for 73 patients that have covid-19 as a response variable and Age, Sodium, Blood pressure as factors. We aimed to study the effect of these factors on the thrombosis for covid-19 patients, as it is clear that the D-Dimmer below 500 means the patient does not suffer from thrombosis but above 500 does suffer, in this situation for such a response of this type Bayesian truncated regression model is an appropriate model to be used to predict thrombosis of the patients that have covid-19, our contribution in this study is using Bayesian truncated regression model to predict the thrombosis for the first time. The study has shown that if a patient's age increases by one year, thrombosis increases by 0.17, also increasing one unit of Na and systolic causes an increase in thrombosis of 0.32 and 0.70, respectively. All three variables are statistically significant because their p-value is less than 0.001.