Accurate prediction of serious infection Especially helpful in cases of the early phase of COVID-19 illness, the flower protein expression could accurately predict hospitalization or death as well as predict who would have a less serious infection. "The method could predict who needed hospitalization with an accuracy of was 78.7 per cent.
With COVID-19 patients who would not have a serious infection, the prediction was accurate at 93.9 per cent," says Associate Professor and Group Leader Kyoung Jae Won, who analyzed the data using machine learning.
In order to analyze the data, the researchers performed a post-mortem examination of the infected lung tissue in deceased COVID-19 patients to determine the flower proteins biological role in