Disease Cluster Analysis in Electronic Health Records: Insights into Mortality and Comorbidity Patterns

Understanding the relationships between diseases is crucial in managing patient health, especially when multiple conditions—known as comorbidities—occur together and increase the risk of poor outcomes. This study uses electronic health records (EHR) to identify clusters of co-occurring diseases associated with higher mortality. We apply hierarchical and k-means clustering methods to find patterns within these disease groups, then use the Apriori algorithm to examine associations between conditions within each cluster. Our analysis reveals comorbidity patterns that impact patient outcomes. These findings provide healthcare professionals with insights for early intervention and personalized treatment plans