Loading...

Data Mining for Rational Use of Drugs

Moradi, Morteza | 2021

273 Viewed
  1. Type of Document: Ph.D. Dissertation
  2. Language: Farsi
  3. Document No: 54177 (01)
  4. University: Sharif University of Technology
  5. Department: Industrial Engineering
  6. Advisor(s): Modarres Yazdi, Mohammad
  7. Abstract:
  8. Prescribing and consuming drugs more than necessary is considered polypharmacy, which is both wasteful and harmful. In this study, an innovative data mining framework is developed for analyzing prescriptions regarding polypharmacy. The approach consists of three main steps: pre-modeling, modeling, and post-modeling. In the first step, after collecting and cleaning the raw data, several novel features are extracted for physicians and patients. In the modeling step, decision trees are applied to generate a set of If-Then rules to detect and describe physicians’ features or patients’ features associated with polypharmacy. A novel approach based on the response surface methodology (RSM) is applied to perform three techniques of feature discretization, feature selection, and parameter tuning simultaneously to consider their possible interactions. In the post-modeling step, the discovered knowledge is visualized to make the results more interpretable and then presented to domain experts to evaluate whether they make sense. The framework has been applied to two real-world datasets of general practitioners’ prescriptions and prescriptions for elderlies. The results showed that there are some interactions among feature discretization, feature selection, and parameter tuning that RSM has identified and the desired values have been determined simultaneously using the desirability function. Also, the extracted rules have been confirmed by the domain experts, which demonstrates the capabilities of the data mining framework in the detection and analysis of polypharmacy. The derived rules can be beneficial for healthcare managers and policy-makers to recognize physicians’ prescribing patterns and take suitable action to support medicines management and develop high-quality prescribing guidelines
  9. Keywords:
  10. Response Surface Methodology ; Feature Selection ; Data Mining ; Parameter Tune ; Rational Drug Design ; Polypharmacy ; Rational Prescribing and Consuming Drug

 Digital Object List

 Bookmark

No TOC