An ontological framework for knowledge modeling and decision support for pharmaceutical product development

Ankur Jain, Purdue University

Abstract

The development of a drug product is a complex and iterative process consisting of selection of dosage form, excipients, processing routes, operating equipments and so on. At each stage, a large number of decisions are made by pharmaceutical scientists and engineers using various forms of knowledge such as experimental data, models, decision trees, heuristics, etc. The objective of this work is to capture and model such knowledge, and use it to develop a decision support system. An ontology-based approach, which provides an explicit and formal description, is used to model the knowledge used in pharmaceutical product development. A knowledge representation model is developed and represented in the form of an ontology, which is called a guideline ontology. The guideline ontology represents decision trees, heuristics and rules in the form of guidelines. Several guidelines are developed based on the knowledge captured from detailed discussions with various academic and industrial experts in the domain. In order to provide decision support, a Java based execution engine is developed for the implementation of guidelines. The execution engine provides decisions based on the knowledge represented in the guidelines and using the information stored in an ontology-based information repository. The execution engine is also integrated with an optimization solver to solve an optimization model, which is developed to narrow down the choices identified by the guidelines. A user interface is developed for the execution engine, which provides all the functionalities to execute the guidelines and to browse the guidelines and results in an intuitive and easy way. This approach provides an open and easy way to create, use and modify the knowledge, and enables the integration of the various forms of knowledge and information. An explicit representation of the knowledge facilitates sharing and retention of the knowledge. As a case study, the formulation of a MDRTB (multi-drug resistant tuberculosis) drug is used to demonstrate the applicability and benefits of the proposed approach. The guidelines are used for recommending the processing route and the excipients to manufacture the drug product as an immediate release solid oral dosage form.

Degree

Ph.D.

Advisors

Reklaitis, Purdue University.

Subject Area

Engineering|Information science

COinS