3.2 Predictions of severity
To assess the similarities and differences of patients with different
severities, principal component analysis (PCA) was applied to reduce the
dimensionality and visualize the patients on a low dimensional space. On
the Fig. 2 biplot, a trajectory from ”non-severe” towards ”non-survived”
via ”severe & survived” patients was observed. It supports that the
blood count parameters and biochemical parameters can potentially
indicate the severity of the patient. Interestingly, the heterogeneity
within the ”non-survived” group is much larger than ”non-severe” group,
suggesting the existence of various reasons for severity. An heatmap of
the overview of changes of laboratory results between three groups is
presented in Fig. S2.