Fig. 3: Subgroup Analysis of Patients with Acute-on-Chronic Liver Failure Categorized by the Lee-2 Model. | Nature Communications

Fig. 3: Subgroup Analysis of Patients with Acute-on-Chronic Liver Failure Categorized by the Lee-2 Model.

From: Unbiased clustering of acute-on-chronic liver failure patients using machine learning in a real-world ICU cohort

Fig. 3: Subgroup Analysis of Patients with Acute-on-Chronic Liver Failure Categorized by the Lee-2 Model.

a t-SNE Visualization of the Clustering Outcome. The green dots represent Cluster 1, and the yellow dots represent Cluster 2. b Mortality Probability of Patients with Acute-on-Chronic Liver Failure at Day 30, by the Lee-2 Model. The red line denotes patients with ACLF in Cluster 1, while the cyan-blue represents patients with ACLF in Cluster 2. P value was computed using log-rank test to determine the difference in mortality probability across the severity clusters. c Mortality Probability of Patients with Acute-on-Chronic Liver Failure at Day 30, by the Number of Organ Failures. The red line denotes patients with ACLF with two organ failures, while the green and blue represent three and four organ failures, respectively. P value was computed using log-rank test to determine the difference in mortality probability across the severity subgroups. d The Top Ten Contributing Variables of Two Clusters Categorized by the Lee-2 Model. The contribution of the variables to the classification was determined by the Kim method. Source data are provided as a Source Data file.

Back to article page