The Application Of Data Mining Technology In Medical Information System

This study aims to learn the structure of medical expenses and the regularity of different populations suffering from diseases through data mining of medical expenses.Two parts were applied in this study.The first is to study the structure of medical expenses by cluster analysis on medical expenses.The second is to determine the association rules between medical expenses and patients’ basic information,including age,gender and type of disease.Firstly,the cluster analysis was carried out based on the type of medical expenses.Four sets,charge separately for diagnosis and treatment,hospitalization,physical examination,and drugs,were grouped and the typical feature vector was obtained by cluster analysis.In view of such a method,any record of any individual could be classified into corresponding set,which provides a basis for the analyses of medical expenses structure.Secondly,association rules with medical expenses were determined using Apriori algorithm.After the selection of item sets and statistic study on frequent item sets,the multi-dimensional association rule among patients’ age,gender,type of disease and medical expense were studied.The results show that patients over 50 years old are more likely to suffer from internal chronic diseases;hospitalization charge predominates in medical expenses in obstetrics and gynecology,etc.Association rules obtained in this study offer a reference for the establishment of medical insurance system and the reduction of medical expenses for target population.