Disease Clusterization Based on Patient Age and Disease Type Using K-Means Clustering

Abstract

This study aims to classify disease types based on patient age using the K-Means Clustering method in order to identify disease distribution patterns at Malahayati Islamic Hospital, Medan. The data used in this research consists of medical record data of patients from October to December 2024, including variables such as age, type of disease, gender, and area of residence. The research stages include data cleaning, data transformation of age and disease attributes into numerical values, and clustering analysis using the K-Means algorithm implemented through RapidMiner software. The clustering results produced three main clusters, representing high, moderate, and low disease prevalence levels. Diseases with the highest prevalence cluster include pregnancy-related cases, pneumonia, acute respiratory infections (ISPA), chronic obstructive pulmonary disease (COPD), and gastroenteritis (GEA), which are predominantly found in adult and elderly age groups. The results indicate that patient age significantly influences disease distribution patterns. This study demonstrates that K-Means Clustering is effective in identifying age-based disease patterns and can serve as a decision-support tool for healthcare planning, resource allocation, and disease prevention strategies in hospital management.