Author/Authors :
Rezaei-Darzi, Ehsan tehran university of medical sciences tums - School of Public Health, Non-communicable Diseases Research Center, Endocrinology and Metabolism Population Science Institute - Department of Epidemiology and Biostatistics, تهران, ايران , Farzadfar, Farshad tehran university of medical sciences tums - Non-Communicable Disease Research Center, Endocrinology and Metabolism Population Science Institute, Endocrinology and Metabolism Research Center, Endocrinology and Metabolism Researsh Institute, تهران, ايران , Hashemi-Meshkini, Amir tehran university of medical sciences tums - Faculty of Pharmacy, Non-communicableDiseases Research Center, Endocrinology and Metabolism PopulationScience Institute - Department of Pharmacoeconomics and Pharmaceutical Administration, تهران, ايران , Navidi, Iman tehran university of medical sciences tums - School of Public Health, Non-communicable Diseases Research Center, Endocrinology and Metabolism Population Science Institute - Department of Epidemiology and Biostatistics, تهران, ايران , Mahmoudi, Mahmoud mashhad university of medical sciences - School of Medicine - Department of Epidemiology and Biostatistics, ايران , Varmaghani, Mehdi tehran university of medical sciences tums - Faculty of Pharmacy, Non-Communicable Diseases Research Center, Endocrinology and Metabolism Population Sciences Institute - Department of Pharmacoeconomics and Pharmaceutical Administration, تهران, ايران , Mehdipour, Parinaz tehran university of medical sciences tums - School of Public Health, Non-Communicable Disease Research Center, Endocrinology and Metabolism Research Population Science Institute - Department of Epidemiology and Biostatistics, تهران, ايران , Soudi Alamdari, Mahsa university of tehran - Faculty of New Sciences and Technologies - Department of Network Science and Technology, تهران, ايران , Soudi Alamdari, Mahsa tehran university of medical sciences tums - Non-communicable Diseases Research Center, Endocrinology and Metabolism PopulationScience Institute, تهران, ايران , Tayefi, Batool tehran university of medical sciences tums - Non-communicable Diseases Research Center, Endocrinology Metabolism Population Sciences Institute, تهران, ايران , Naderimagham, Shohreh tehran university of medical sciences tums - Non-communicable Diseases Research Center, Endocrinology and Metabolism Population Sciences Institute, Endocrinology and Metabolism Research Center, Endocrinology and Metabolism Clinical Sciences Institute, تهران, ايران , Soleymani PharmD, Fatemeh tehran university of medical sciences tums - Faculty of Pharmacy - Department of Pharmacoeconomics and Pharmaceutical Administration, تهران, ايران , Mesdaghinia, Alireza tehran university of medical sciences tums - School of Public Health and Institute of Public Health Research - Department of Environmental Health Engineering, تهران, ايران , Delavari, Alireza tehran university of medical sciences tums - Digestive Oncology Research Center,Digestive Disease Research Institute, Shariati Hospital, تهران, ايران , Mohammad, Kazem tehran university of medical sciences tums - School of Public Health - Department of Epidemiology and Biostatistics, تهران, ايران
Abstract :
BACKGROUND: This study aimed to evaluate and compare the prediction accuracy of two data mining techniques, including decision tree and neural network models in labeling diagnosis to gastrointestinal prescriptions in Iran. METHODS: This study was conducted in three phases: data preparation, training phase, and testing phase. A sample from a database consisting of 23 million pharmacy insurance claim records, from 2004 to 2011 was used, in which a total of 330 prescriptions were assessed and used to train and test the models simultaneously. In the training phase, the selected prescriptions were assessed by both a physician and a pharmacist separately and assigned a diagnosis. To test the performance of each model, a k-fold stratified cross validation was conducted in addition to measuring their sensitivity and specificity. RESULT: Generally, two methods had very similar accuracies. Considering the weighted average of true positive rate (sensitivity) and true negative rate (specificity), the decision tree had slightly higher accuracy in its ability for correct classification (83.3% and 96% versus 80.3% and 95.1%, respectively). However, when the weighted average of ROC area (AUC between each class and all other classes) was measured, the ANN displayed higher accuracies in predicting the diagnosis (93.8% compared with 90.6%). CONCLUSION: According to the result of this study, artificial neural network and decision tree model represent similar accuracy in labeling diagnosis to GI prescription.
