Leveraging Text-Mining Techniques On Electronic Medical Records to Analyze National Drug-insured Medication Use

Processing electronic medical record (EMR) data has become a common practice among scientists for extracting valuable insights and studying diseases. Given the large volumes of text in EMRs, efficient computerized text-mining techniques are necessary. As academics, we recognize that drug-used analysis from EMR Indonesia is currently limited. This study focuses on obtaining meaningful to make positive recommendations hospitals. The proposed method uses pattern-based Regular Expressions (regex) extract drug names Levenshtein distance algorithm check their compatibility. We developed pattern based analyzing data. extracted were compared list selected drugs (National Drug-Insured/Fornas) required must be provided at healthcare facilities Indonesia. threshold was set two decide whether belonged nationally drug-insured or not. Only about 11.09 – 16.11% medications given by doctors listed Fornas list. Between 2019 2021, there an inaccuracy writing prescriptions drugs, with as many 57.53% 63.21% being written incorrectly. results this indicate promising potential implementation Ministry Health Indonesia, precision rate 97.07%.