Publication: Diabetes dose titration identification model
Issued Date
2016-02-04
Resource Type
Other identifier(s)
2-s2.0-84969219522
Rights
Mahidol University
Rights Holder(s)
SCOPUS
Bibliographic Citation
BMEiCON 2015 - 8th Biomedical Engineering International Conference. (2016)
Suggested Citation
Ratchanee Kaewthai, Sotarat Thammaboosadee, Supaporn Kiattisin Diabetes dose titration identification model. BMEiCON 2015 - 8th Biomedical Engineering International Conference. (2016). doi:10.1109/BMEiCON.2015.7399557 Retrieved from: https://repository.li.mahidol.ac.th/handle/123456789/40621
Research Projects
Organizational Units
Authors
Journal Issue
Thesis
Title
Diabetes dose titration identification model
Other Contributor(s)
Abstract
© 2015 IEEE. Diabetes is a chronic disease that requires continuous treatment throughout lifespan and increased risk opportunity of developing a number of serious health problems, which are high treatment cost. Admitted diabetes inpatients should receive the appropriate treatment in order to reduce rating of severe complications and premature death. This paper aims to develop the classification model for diabetic medication adjustment based on historical medical record of diabetic inpatients by applying three algorithms; Decision Tree, Naïve Bayes and Artificial neural network By comparison of the results of each method, Decision Tree is outperformed than others for Independent Dose Titration Model (IDT) dataset and Artificial Neural Network algorithm generated model with high accuracy and ROC Curve for Historical Dose Titration Model (HDT) dataset. The results of this paper could support the decision making in medication adjustment of diabetes inpatients, particularly type-2 diabetes inpatients.