Clustering the Whole-Person Health Data to Predict Liver Transplant Survival

Stud Health Technol Inform. 2016:225:382-6.

Abstract

This study aims to discover groups (clusters) of patient who share whole-person characteristics. An unsupervised clustering analysis using a hierarchical agglomerative approach was applied to identify meaningful groups of patient characteristics. Results showed that is possible to identify clusters that have similar patient characteristics, and that these characteristics may be associated with survival.

MeSH terms

  • Cluster Analysis
  • Female
  • Humans
  • Liver Transplantation / mortality*
  • Liver Transplantation / statistics & numerical data
  • Logistic Models
  • Male
  • Middle Aged
  • Risk Factors
  • Survival Analysis