Figure 1
From: Identification of the robust predictor for sepsis based on clustering analysis
Three subgroups established by K-means clustering. (A) Three subgroups were characterized based on seven sepsis-related variables using one sample t-test. The brown box plot is about all subjects including healthy control and sepsis patients. Black middle, upper margin, and max lines in each boxplot indicate means, one standard deviation (SD), and 2-SD levels. (B) Beta-coefficients were measured from multivariate logistic regression. Brown bar was made by logistic regression for all subjects. Other bars (yellow, green, red) were curated by logistic regression for each cluster-specific subjects. A red asterisk indicates p-value less than 0.001.
