Early Prediction of COVID-19 Patient Survival by Blood Plasma Using Machine Learning

The Coronavirus Disease 2019 (COVID-19) pandemic has severely disrupted the global healthcare and medical system. Although COVID-19 is no longer considered a public health emergency of international concern, it can still cause many infections and even life-threatening conditions. This study aims to reveal novel potential biomarkers of mortality and identify associated mechanisms of death caused by COVID-19 using machine learning approaches to re-analyze metabolomics data. We found that the combination of NG,NG-Dimethyl-L-arginine, 4-Coumaryl alcohol, Pyridoxamine, N1,N12-Diacetylspermine, Coniferyl alcohol, 1-Phosphatidy1-1D-myo-inositol 3-phosphate and sn-Glycero-3-phosphocholine is an effective biomarker for survival prediction of severe COVID-19 patients. These metabolites suggest potential immunomodulatory therapeutic strategies for the treatment of COVID-19.