Disease detection in medical prescriptions using data mining tools

Abstract

Prevalence of communicable and non-communicable diseases is one of the most important categories of epidemiological data that is used for interpreting health status of communities. This study is aimed to calculate the prevalence of outpatient diseases through characterization of outpatient prescriptions. The data used in this study is collected from 1412 prescriptions of various diseases and we have focused on the identification of ten diseases. In this study data mining tools is used to identify diseases related to each prescription. Then we have compared the performance of these methods with a Naïve method. The results indicate that implementation of data mining algorithms has a good performance in characterization of outpatient diseases. These results can help to choose the appropriate method for classification of prescriptions in larger scales.